The $10K Mistake Every Service Provider Makes With AI

Today I'm going to talk about the biggest mistakes that business owners are using with AI and what you can do to fix it. All right, guys, welcome back to the show. We just held a in-person mastermind this past week here in Austin, Texas. And we actually were talking about and unveiled this specific diagram that I think is going to clear a lot of things up for you.

And I'm going to talk to you about mistakes along the way and why this model works for you specifically as a business owner versus maybe other things you're hearing about. Every day there's people talking about new ways to achieve things with AI. And I'm going to talk about the way we're using it and the structure that I think you can follow and adhere to as a general framework that helps you maximize your outputs, get results specifically with your marketing that actually feel like it's your voice, but also have your company run the way it could and should utilizing AI and layering in your tools together.

Okay, so let's go ahead and jump in here. I think the first and foremost thing that I want to talk about is a lot of people look at AI as a replacement. We're looking at AI as an enhancement. And what we're what we're personally seeing is people, when you have that fusion between you, the thought leader, you with your intuition, you with your expertise that you bring into it, lived experience, plus the power that AI can bring to your work.

That's the power combination. And in fact, we're taking it so far as to help people like you would lend a computer or a company computer to an employee. We're actually having other team members of ours also be equipped with this exact model. And I'll talk about that a little bit more here as we go.

But the core to remember here is that it's you at the center first, right? Nothing you cannot replace that. The soul of the business really runs through you, specifically as a small business owner. And so it's really important that you keep you at the center and then build around that.

No different than what you would do in a physical world. You're building the company around you so that you can show up and fulfill the dream that you had for your small business. So that's where it starts. Next is establishing your thought partner.

Now, everybody has their preference here. Mine particularly is on Claude, but others, it might be ChatGPT, it might be Gemini, it might be Perplexity. There's plenty of language models that maybe fit your needs better than most. For me, being more right-brained in ADD, the precision that Claude provides and the detail that it gives me, it helps with the execution aspect.

The step three handoff for me from ideation and conception of idea into actually handoff into an execution layer. And so I find Claude to be the most effective for me specifically. But whatever thought partner you establish, just make sure that it's first and foremost, it's running point with you, right? And so it serves as just that.

It's your thought partner. Now there's an additional layer that we've added, and this is what we talked about at our mastermind. So we build a product called Spark. You'll hear us talk about that a lot.

It's actually an acronym, and we'll talk about that later. But what it effectively does, it allows you to have a marketing co-pilot. And so what it does is it takes your thought partner and it gives it our frameworks, our methodologies, but layered into your actual thought partner. So when you're inside of a session in Claude, for example, you can actually say, Hey, can I use the Spark frameworks to help with a sales page?

Can I use the Spark frameworks to write some ad copy? And it Claude will do that on its own, but the ability to reference marketing frameworks that we've worked over 10 years to create, the outputs are clear and defined. So there's three layers there that are really powerful already. It's one, it's going to be able to capture your voice because you've trained your thought partner to do that.

Two, it's the power of the AI model to begin with, which obviously, if you've used it, you know how significant that is. And then three, layering into specialized knowledge, in this case, our marketing frameworks. And so your outputs already, with that triage of attacking whatever you're trying to build there, you're already equipping yourself with the right framework, even if you stopped right here. Now, what a lot of people are doing is as an example, is they're not establishing that thought partner.

And what I mean by that is they're going into Claude or chat or whatever they use, and they just have these marathon threads and they don't live inside of a project area. They don't have context. You haven't set your account level settings and preferences. You haven't given it files that are relative to what you're trying to get it to do.

So all of these things with establishing your thought partner, we'll talk about that in another episode. I just want to give you the concept here of what this needs to look like. So if you haven't done that yet, what you really want to do next is make sure that you're thinking about setting up something inside of a project. And that is what I'm talking about when I say thought partner.

And so obviously, a cold thread that's not trained is going to provide less of an output as something that has context about you, context about your business, how you do things, who you serve, the messaging, how you communicate, your team members, your goals for the year, your business plan, like all of those things. Imagine if you were working with a coach that didn't have that context about you. The capacity is limited, and what can be provided is limited because it doesn't have that additional layer of context.

And so these language models have the ability to encompass a lot of and capture a lot of context for you to serve you in the ways that you need it to serve you. But in order to make it a thought partner, that's step one is to make sure that you're actually taking those appropriate steps. And if you're interested in doing that, I'll be doing that in a future episode. You've probably already seen episodes of mine where I actually talked to my specific thought partner here on the podcast on the YouTube channel.

And so I'll link that as well, where I've actually pulled Tarvis, my business thought leader, into the show with me and we've had conversations in real time in the show. So a lot of people enjoyed that. If you want to check that out next when you leave here. So this is the layer two and two B for us.

So 2A and 2B is what this looks like. So it's us at the center. It's 2A is the thought partner that has that rich context, and then 2B is the actual frameworks that it taps into, and that that serves us well, right? So we don't have to go through a course and then remember all the frameworks that are already built in to be able to utilize and deploy right there.

Okay, next. And I say this term loosely. I know there's gonna be some technical folks that kind of get on me about this one in the comments, but automations. And so what I effectively mean by that, and some things aren't technically defined by an automation, but it really means is where are you sending things from your thought partner that actually help your organization run, right?

Your ecosystem of tools probably live in silos right now. My guess is you have a lot of things happening on different platforms, and that data just sits there isolated, or you have a lot of tools that don't communicate. And so what we really want to create here is it really should probably be called connections instead of automations, but it's establishing those connections. It's making sure that your tools all have a central hub, but they're all connected and you have an ecosystem where, like for me, for example, I will push updates directly from my thought partner into ClickUp.

I will push tasks to people. I'll have it check in and help me build our annual plan and our quarterly rocks and then tell me where we're at on those things. Give me an update of where we are so that I'm not having to keep track of those things on a daily basis. There's a litany of things that you can do once you establish that secondary layer here of having those automations as the place where you connect your tools and house what you need to have housed.

And so we're gonna talk about what that looks like in a practical sense here in a second. Just know for now that we want to make sure, think of this like you and your thought partner doing something on your own, and then no one else is included. And this makes sure that everybody can be included and all your tools are speaking together as at least as much as they possibly can. Okay, and then the last layer, and you've probably heard a lot of people talk about this because this is like the buzzword right now, but it's agents.

And so agentic AI is very, very powerful. We won't talk about the three layers or the three different types of agents right now, but for your reference, what they are their server side, browser side, and desktop side agents. And so these are things that are doing ambient tasks. There's a real risk with just deploying agents if they don't have a place that they're supposed to go.

And so if you don't know what an agent is, a lot of people think that they just kind of like look over your business and do everything for you when you're not there. It's not entirely real, although maybe one day it will be. Right now they're very good. I give the reference of think of them like virtual assistants, but highly effective and can work 24 hours a day.

They're triggered by an action, either time-based action, meaning like a certain time elapses, like every Sunday something happens, or a trigger of an actual action that actually triggers it to do something else. So, like dropping something in a click-up card for us will trigger an action. Every Sunday, one of our agents will pull a report from our convert kit, from our YouTube, from our emails, and then that report surfaces to us inside of ClickUp and then services all the way to the top where we'll give that to our thought partner to help us analyze what are the meaningful takeaways for that week.

So it's bi-directional, and you'll see here I've got arrows going both ways because all of this is bi-directional. So it goes from conception of idea with a thought partner, utilizing frameworks, pushing that into environments where all of our tools are connected, right? And then whatever agents need to be doing and deploying off of that, for example, could be creating a YouTube video through our framework, right? Making sure that's really refined with the transcript that we're or the script that we're going to utilize, pushing that into D script because that's a connected tool for us, or into clickup for our team to be able to reference, right?

In this case, for us, the transcript actually triggers an agent that allows it to pull meaningful clips where it creates 16 pieces of content from your one transcript from your video. And so all of that works together. In addition to that, we'll drop another, it'll pull some of the data from YouTube. So let's say you're pushing that to YouTube biodirectionally on the way back, it could be pulling the data from that video or all of our videos to provide a report all the way back through ClickUp, all the way to surface back to myself and my thought partner.

So we can review those things. And I get that report every Sunday night. It's just a full company overview, what's going on, what are the data sets telling us. I don't want to sit there and look at that personally.

I don't know if you have the desire to do that. But what's great is that if you have your thought partner, it can do that like that and give you the meaningful takeaways and provide a summary. If you have like a meeting on Monday mornings with your team, great, it can put it in a format that you can present to your team. So that becomes really powerful.

So hopefully you see here how this works layered, you're right, all the way starting with you, always starting with you, and then amplifying out to your team and then having that like agentic work there on that outer ring. And so the reason I want to make sure you guys understand, at least for small businesses, and maybe this isn't the case for everybody, but it's really important that your agents are spawning from somewhere, right? That they're connected in meaningful ways. And so I want you to think about a time where you may have had a virtual assistant and you're like, I don't really know what they did this week.

I know they I think they worked, but I don't really know what's happening in the digital sense in this regard, using that same metaphor, is they're called orphan agents, is they're doing work, but nobody sees the output. Stuff's just happening, right? And so you're like, well, I think the agents are working, I think they're doing what they're supposed to do. And so again, we want to have a connected ecosystem, just like you would have with a company, just like you need to keep tabs on what's going on, just like you provide a place for people to work that where things can be streamlined and communications can be organized.

And so that's really what's happening here is making sure that you don't have orphan agents that are just outdoing things that aren't surfacing back to your team or to you in some meaningful way where it's being put into action. And so a lot of people are gonna fall victim to this. And I think it's really important that you make sure that you understand what an agent is, what it does. And even when agents become more powerful and can start making decisions on their own, like helping you manage your meta ads, for example, you still want to have guardrails in place and you obviously still want that data to surface to you.

You don't want to find out that you ran ads for a month. It was making decisions on your behalf, and those weren't decisions that you're even aware of that were happening. And so I think it can actually become kind of like a cascade of bad decisions if you're not careful. And so always having those guardrails in place, I think is still going to serve in your best interest.

So here's the metaphor I gave to the folks that attended our mastermind, and maybe it resonates with you as well. But think a metaphor of the body, right? So, like we think with our brain, right? Then we connect through our central nervous system, and then we execute with our hands and feet.

And so this is you and your thought partner, the think portion. The connections are all of your tools, your click up environment, or whatever project management you use, everything's running through here. And then in an ideal world for your business, the execution is the agentic layer. And so that's the hands and the feet.

Those have to send signals all the way through the central nervous system and then back. So if your hand receives pain, well, it sends that back to your brain through your central nervous system, and you're a, oh, don't do that again. Right. And so that's how I like to think about it, making sure that we have a really refined, closed system where everything's speaking to one another and we're not deploying things just because they're cool, which is awesome.

And it's fun to play around with things, but making sure, okay, this is great, but where does it surface and what is it providing me? And are those meaningful things? And does it resurface back to where we can use that as something that we're going to either further iterate or implement in some way or at least run through our thought partner? So really important on that level as well.

So let me show you what this looks like practically, so you kind of have an idea of that. But think here's you and your thought partner. This is called MCP Access. It's model context protocol.

You don't need to know what that means. If you are a Tron fan, I'm wearing my Flynn shirt, you probably have a tough time with that because you want to say master control program, which that would be cool, but it is what it is. So model context protocol is what this means. Think of this like the USB C of the digital world, right?

And so imagine if you were wanting to do something and you just want to connect your USB C. I want to connect this phone that I'm holding to my computer because I want to transfer some photos over. Well, that cord allows that connection to happen. And so for us, this is our high-level architecture here.

We go directly into ClickUp. That's the central nervous system for us, or the core of the central nervous system. And then I've mentioned we're connecting various things into that. So there's agents operating inside of here that are pulling data, ClickUp agents in this case, pulling data from YouTube.

We're able to actually connect into right now our inbox. So we get support tickets. We have an agent in ClickUp that actually pulls that in and actually drafts a response for us. Our SAM card, so our sales data gets pulled in, our Google Analytics get pulled in.

So all those things are happening inside of this ClickUp environment. Again, I can push tasks, I can push updates directly into ClickUp from the thought partner. I can pull things out. So if I want to know what the team worked on this week, and if they're leaving their daily reports, then it'll tell me what they worked on this week and time allocation-wise, where we can optimize.

Or if we feel like they didn't do enough that week or it wasn't fully dialed in, or we need to make some pivots because we didn't assign tasks well, then that would probably surface to me as well because it'll say, Hey, I'm going to flag something here because my thought partner's trained on knowing the team members and what they're responsible for. So we can make some key critical decisions. But you can see here with having these, just these four that I've referenced, including ClickUp being a fifth, can provide tremendous value to me and my clawed thought partner named Tarvis.

And so all of that eventually surfaces back to me. And then I can make the decisions I need to make to run the company the way it needs to be run. And so there are some other agentics behaviors, agentic behavior that's happening, other agents that are deployed. And so, like our own Spark agents could be an example of that where I need that agent to do some research for me.

For example, new leads that come in. I have an agent that actually researches that based off of what we look for, a potential lead that comes in to click up. It saves it under our little sales pipeline. It actually researches that person and it gives me a score, how much of a fit they are for me.

So it'll track their URL, it'll look them up online, it'll research their business. And that way that saves me probably 30 minutes of time that I don't have to spend to go out and research a new lead. And that enables me to run ads aggressively because imagine all that's being done. Those high-ticket quality leads are coming through, and it already drafts an email for me.

It sends them a text message that's customized, not just a random text message, but all of that spawns from either click up or directly from that agent that I have trained customized for my specific needs. And then it allows us to amplify or accelerate that process tremendously. So just imagine in that one particular example how much of an advantage that you have over somebody else that's not running something like that, that's literally looking through every lead manually or is not sending a customized text message like what we're sending.

And so in that one particular example, that agent is pulling that in, doing that research, creating that card, drafting up that custom text message, actually drafting an email for me to send to that specific person. And then they're already on the schedule. So just sending the follow-ups as far as getting on that call with me. And so that process alone probably goes from maybe several hours down to 30 seconds, realistically, and it allows me to stay in the work that I am most capable of doing.

And so that's just one example of many. And we've got several agents that are doing various things, but specialized for our specific needs. And inside of the Spark platform that we built, our clients have access to utilize those as well. So they can customize and deploy for their specific needs.

It would be doing lead research, but through the lens of what a good lead looks like for them. So that's an example of what this could look like. And so this was a fun image that my wife Kim created. But when you have your generic AI, this is what it looks like.

And when you have your thought partner plus Spark in the ecosystem that we built, it's a mega superpower ecosystem for yourself. So hopefully this video answered a lot of questions for you. I know these are brand new concepts, and so everybody has their way of doing things. You that you might fundamentally disagree with the way we're doing it.

But for us as small business owners, it's what we found to work best for us. If you did want to explore some of the concepts that we discussed here, or you felt like there might be some questions that you have that are lingering questions, you want to make this work for you, or you're using similar tools and you want to know how do I make that work for me? Like I want that to have what my business life looks like. There is a discovery call link you can grab below.

It's 100% free for the time being. You can grab that and book a call with me directly, me and my wife Kim, and we'll talk it over with you and see if it's a good fit for you to explore what AI integration would look like for you and maybe just answer some of your questions. Maybe taking the next step, next steps into working with us doesn't make sense, but just getting some pointers on some next steps maybe is where you're at right now, and we're totally good to meet you there as well.

So you can check that out in the description below. Don't forget, while you guys are here, if you want to hear more about this kind of thing, you want to see more of what we're doing with AI and how we're connecting our tools from a marketing perspective for small businesses specifically, make sure you guys like this video. Helps with the algorithm big time. Subscribe so you see more of this and make sure you hit the bell for notifications.

A lot of people subscribe to things and then they just never see that channel again. And so if this is something you want to hear about, be like, I really don't want to miss these things. Make sure you hit that bell so you get notified about that. And then if you would in the comments below, but let me know what you want to hear me talk about next.

Was there something that we talked about today you want me to go deeper into? This was a lot a little bit more conceptual and high level, so that we could go deeper into the trenches on that. But I want this channel to make sense for you guys and you guys get what you need out of it. And so if there's things you want me to cover, you can drop that in the comments below as well, or what your biggest takeaway was from this particular video.

So that's all I had for you today, and I'll see you guys in the next one.