Intro And What We Solve

Dave 0:01

Welcome to the Business Crypto Podcast. We're here to share real life insights, practical strategies, and the added lessons that we have learned from our own state. So whether you need help with operations, accountability, financial knowledge, details, or maybe just getting your mindset right, we are here and it is the right spot for no fluff conversations and tools to help work. So grab your favorite cup of Joe. Let's jump into the show. Let's jump into the show, ladies and gentlemen. It is another Friday morning, which means you're here with the Business Unscripted Podcast, or at least you're maybe starting to, you know, do all that fun rolling in and everything like that. But for those of you who don't know who we are, it's my name's Dave Warden, and I'm here with one of my business partners, Dwayne Bernhagen. And we do this show every single Friday for you, the business owner, who maybe is you know looking for somebody who can share some advice, some strategy. Maybe you can learn from our mistakes, because you know, Tarn, me and you, we probably make a few. We're not perfect, but you can learn from our mistakes, learn from other things that we're helping our clients with in business as well. I've been in business for over you know 10 years now, and it's my third venture, you know. So we lived, we learned, I've failed a few times, but that's okay. But 20 years in leadership and all that fun stuff, right? So it's you're you're in the right spot if you're a business owner and you want some advice. So we you know hope you enjoy the show. So let's jump in, Dwarne. Brother, welcome back. I know last week you were doing all the uh the fun legal things that sometimes you have to do when you're going to a big change, like we've talked about. So, how

Corporate Restructure And Contract Fallout

Dave 1:33

are things?

Duarne 1:33

Absolutely. How are things? Well, look, things have been very hectic, and I'm looking forward to a mostly quiet weekend. We had so you're right, we we did a company change, it was a core as a corporate restructure. We changed from one type of business to a different type of business registration on paper, and there was lots of legal benefits for doing so, but also created a lot of work. Uh, whilst we absorbed all of our brands and with the new company, what tended to, and this is something we'd been working on since January, we finally uh bit the bullet and decided to do the final push. Lots and lots of stuff to take care of. New contracts for clients, new contracts for staff, new bank account, lots of legal-related matters to consult with lawyers because certain situations occurred during the process and that we didn't foresee. And as well planned as it could have been, shit still hit the fan in a couple of areas. So it's okay.

Dave 2:48

All right, all right. The proverbial, you know, shit hits the fan sort of comment. What does shit hit the fan look like for you? Because I'm pretty sure everybody's got different shit, and their fan might be a little bit different. Maybe we're all power high powered. So, uh, how much shit got on the walls and all over the place? As much as you want to share, of course.

Duarne 3:06

Yeah, yeah, yeah. Look like the long and the short of it is with the new contract re-signing for the staffing side of things, we had every person, so over 25 staff signed new agreements with the transition, with the exception of one. And one had a hold out typical, but typical, yeah, held out. They they put up reasons, and yeah, look, we we spoke to our lawyer about it, and the lawyer's basically given us some good advice. We still need to push forward with that, so I can't share too much about what that advice is at this point. Okay. It's still got to be actioned, it's still a work in progress, but still open one. Hey, buddy.

When Systems Exist But Aren’t Used

Duarne 3:52

One thing that I can share is that this person, this employee was actually assigned to a client, and the client learned a very valuable lesson, and that was yes, if you can have systems in place to monitor and review your staff, but if you're not using those systems and checking them, then things are gonna slip through the cracks. And his particular instance, 12 months worth of slipping through the cracks when he did his back checking and reviewing to find that underperformance by about 50% of the output that he was that was needed, so I think was the case.

Dave 4:38

And if you were a concern that I think a lot I've heard a lot from you know business owners who are questioning, does the VA make sense for me? And it's that particular situation, is I can't control what they're doing, right? Or I don't know if you know the output that they say they're doing and the hours they're spending are the hours that they're doing. So obviously, your your experiences right now where the client obviously, you know, and ultimately it is the client responsibility, you know. We're paying for an output to be given to us, and so you know, it'd be very difficult for you as a you know, like a company owner that's hiring these VAs to just have everybody in one place, which a lot of times you don't. So what advice can you give, right? To without giving away too much information of what this person was doing on absolutely, yeah, yeah.

Duarne 5:35

Well, I mean, look, yeah, in this particular system, this particular instance, systems were in place, but they were never used due to circumstances which caused the client to be lulled into a false sense of security. And okay, I'll explain what I mean by that in a moment.

Dave 5:57

Yeah, yeah. I I feel like there's some undertones and some innuendos going on in there, but uh yeah, yeah, absolutely.

Duarne 6:05

Go ahead. It was it was rather intriguing because I couldn't understand it when I was going through doing a review with the staff, sorry, with the client. He was looking through his systems, which very, very easily report all of the output from the his um staff member. Okay, and he hadn't looked at it for 12 months.

Dave 6:30

Well, and I couldn't know he had he had a report.

Duarne 6:34

He had it constantly reporting, and he never looked at it. Well, he never checked for the he never checked the results, so he made the assumption that everything was okay. So the systems were in place, okay, but they needed a manual process to check them from his side, just to basically as a sanity check, right? And it seems really obvious, but 50% reduction in output when he's handling the billing, and I would have expected to be something got seen and noted, but when I pushed the point with my client, what came back was he had had very good feedback of the staff members' customer service from multiple clients of his, so that made him feel like everything was working and everything was going well. There was billable hours going out, but he didn't realize they weren't as high as they should have been. And on top of that, he was getting contacted 10 to 20 times a day with I'm just updating you. Hey, just touching base to tell you that I've just done this on a ever almost every day. So he was lulled into a overcommunication, yeah, yes, overcommunication from his staff member, which made him feel like, oh, everything's happening because I'm getting lots of communication back. So he felt comfortable enough not to check the actual logs, which would have shown a different story and forced him to take a look. Now, when this all happened, of course, we had to talk to our client and explain it. In this instance, the staff member circumvented our process and went directly to the client to complain about something about the contract, which a restructure internally had nothing to do with the end user client, so therefore wasn't communicated because it was not making any difference to their side of things at that point. That would have come later.

Dave 9:13

But I want to think like hit something really quick before we get too far away from it. And so, and it's just this idea that the system was there, your client was probably overwhelmed doing a ton of other things, so they just assumed that they didn't hear anything negative, so everything was fine. And I think that's the part that we want to hit on is that you never want to just assume just because you haven't heard anything, or you're being overcommunicated that things are fine. And this is where you know we talk about AI in this podcast a lot because we're both into it a lot. This is where kind of evolving the system to include some sort of AI review, right? Where now you have the AI, you know, maybe at the end of every week, kind of looks at all the hours and trends or picks up things that are out of the ordinary or don't seem to be in line on where they should be, right? And the expectations, and you set those expectations in your AI system. And then when something does stand out as the abnormally anomaly, then it can either ping you through an email, it can ping you through you know a push notification, depending on how you set it up. But I think that's a piece that you don't want to miss now is that yes, you can have these systems, and in the in six months ago, a lot of these systems still required a lot of manual review at the end of kind of seeing it and where you're at. But now the beauty of where we're at today in AI is that you can then set up that that next level of, hey, now I'm gonna take this manual review, take everything that I want to look at in terms of KPIs and metrics and trending, and turn it into an AI-run automation that checks all of that information for me and then just reports out when there is an anomaly and things need to be looked into and checked.

Duarne 10:54

So crazy enough though, in this particular instance, Dave, is a simple report

Automations And AI For KPI Alerts

Duarne 11:02

could have been created on an automation with the system that the client was using. That's and it could have simply just came up and be alerted that there was a negative based on what the KPI threshold was. And it could have been sent out on a weekly, monthly, quarterly, whatever. And this was something that could have been set up, and that's just a simple automation capability without even going into the AI side to at least tell you what something is, something's wrong, and to give you a chance to go and dig deeper or do whatever you need to to make you know to investigate that. So it doesn't always have to be AI in this case either. It could just be automation to generate a report and send it, right? We have the tools and a lot of systems support either either solution, right?

Dave 11:51

The beauty of the beauty of that too, I guess when when when I when I think of AI is just AI is everything now because there's some sort of like even automation is some sort of artificial process that you put together, and AI is kind of that same thing, it's changing things together. But you know, building that you hit the key point, right? People mentioned that you have KPIs and you're not aware when you have variances from the KPIs because it's not in your face, right? It's not monitored anywhere. You know, you may have the data someplace, and you're like, Oh, yeah, we track that. Okay, but how do you use that data that you're tracking? Yeah, if you don't have the system, yeah. Who's looking at it? How are they getting the data? You know, and how informative is the data, right? Because there's a difference, and I'll use financing as an example, right? Accounting, you know, there's a very big difference of me giving a high detailed variance report with KPIs and spark lines and uh you know, plus or minuses on revenue goals or gross profit percentages and all that kind of stuff, versus me printing out a general ledger that has all the transactions and giving it to a business owner. Which one do you think is going to be more informative in terms of ease of making a decision and knowledge of where their business is at? Option A or option B?

Duarne 13:08

Um, option A.

Dave 13:09

Option A, right? Option B, of course, is going to be more informative because you have more information at your fingertips. But if you don't have the time to go through it, you don't have the time to like parse it.

Duarne 13:19

Well, let's go back a step. It's not even about the time in some cases, depending on what your systems and what the person's doing, you may not even have it may not be time related, it may be skill set related. You may not have the skill set to understand everything without it being pointed out in obvious detail to you. And that's where setting up the right automation, the right summary. This is one of the things that we put together with all of the AI assistance we have now. We take SEO reporting and pay-per-click reporting and then turn that into a one, two-page summarized understanding for our customers. So we can take that 15, 20 page report, still include it if you want to look at it, but we'll just we'll give you the summary overview so you can quickly see am I winning or am I losing? If I'm losing, how bad am I losing? If I'm winning, how good am I winning?

Dave 14:12

And it's a fine balance too, because when you're when you're like a service provider, you feel like you have to give them more information that you have to feel like you have to give them more in order to actually like like justify your service, right? So your value. Yeah, your value when in reality your only goal is to meet their needs, right? What is important to them, yeah, give advice, right? Say, well, hey, maybe we should also look at this, I'll include that in the report, but don't overwhelm them because remember, they're busy doing what they're good at. You know, I think of one client of mine, you know, a highly successful, you know, real estate agent. You know, and for me, you know, to me, it was, you know, initially, hey, we're gonna get you some financial statements, we're gonna, you know, give you the insights that you need. And the more and more I work with them, I realize, hey, that we're gonna pair this back a little bit. We're just gonna give you the information that matters. Plus, we're gonna help you kind of manage better, you know. So he's, for example, he showed me a spreadsheet that he has that you know he's tracking and working with his team. I'm like, give me that. We'll just turn it into a live dashboard, we'll let even your team be able to log in and you guys can actually manage it more effectively. And so those are the things that are valued. Like, how do you take your client's time depending on where you're at? And again, it's like and I said this yesterday in a presentation panel discussion I was on. I understand nobody, not nobody, there's not gonna be many people that have the passion around AI that you and I do, okay? To really dive in, do it 13 hours a day, and be willing to you know use it in a way that's gonna help. But there are gonna be the people that have interest in it that are like, oh, I know I need to do something, but I just can't do it all. Great, that's why, right, there's people like you and I, because we can then take it to that next level. You can just be open to the fact of, hey, yeah, I want to start doing something with AI in my business. I don't know what it is, but let's have a conversation and just dive in and try it and just test it. You know, I I just asked, I was asked a question like, how did you get where you are today from where you you were with no technical background in under a year? And I'm like, because I I was willing to make mistakes, I was willing to learn.

Duarne 16:25

And failure is learning.

Dave 16:27

Yep. I was willing to learn to fail. I was willing to go multiple versions of something in order to make it right. But and don't get me wrong, like there's a lot of frustrated back and forth with with my you know, Claude or my you know open AI or whatever, with you know, maybe some F-bonds in there, you know, dropped, and and hopefully when they take over the world, they don't come after me.

SPEAKER_01 16:48

You know, like really I'm like, man, what on the F did you do that?

Pricing Failure And Learning Curves

SPEAKER_01 17:29

I never thought you could do that.

Dave 17:31

Yeah, exactly. And it's just like there's so many frustrated times of but you have to be willing to just test and try and get in there and just get started.

Duarne 17:41

Make mistakes, learn from those mistakes. It's like anything in life. Uh it's really interesting. Like, I was watching something on YouTube yesterday. I was having a bit of downtime over lunch and I was watching, and there was a Bali-based Indonesian workshop that makes reproduction.

Dave 18:01

Do you have any questions, by the way, while you're here? Like I see the viewers, like drop them down below. You're always open to have questions, and we'll answer them live on the show. But sorry, go ahead.

Duarne 18:10

Yeah, they make reproduction versions of classic European cars. Aston Martin's, Doll Wing style, Mercedes, all that sort of cool stuff, and a couple of Japanese, like the was it the 2000 GT 2000, which like 200 were made of the Toyota's. Oh, that's a while ago.

Dave 18:30

Yeah, wow.

Duarne 18:30

They go back to like like Fast and Furious, the Drew Man, like with like we're talking like you can't buy parts for these cars. So they manufacture every single component in in Indonesia and Bali. They've got six molds made out of wood, and everything's been made from diagrams. Everything's been made from blueprints have been able to secure, from photos that they've had printed on tarpaulines, so they can blow up and look at them and try and get their measurements and everything. They make they even create their own glass, they have that they do their own components for everything. It's like 100,000 components across the five cars that they build, and they use drivetrains from vehicles that are available in country. And the whole reason they do is because you can't import vehicles over a certain age into the country in Indonesia. So this is their workaround to make these classic cars for the local enthusiasts who want them. And they do beautiful, beautiful replica vehicles. And the reason I was saying this is because one of the questions that was asked by the foreman who assistant to the owner was do the people who come in here know how to do the job? No, because there's a huge car enthusiast group in this country, but nobody knows. So we train each and every one of them. And because there's so many components and everything is handmade, like we take into consideration the cost of failure on every single vehicle. There is a cost of failure, and we factor that in with our staff. Our staff will fail, and that's okay. We expect them to fail from time to time. We expect them to mess up a component or a part, we expect to replace that part and have to build it again. We're gonna have to recover the time and the material to do that, and we're okay with that. That's factored into the price to the client. And when I was hearing that, I was like, it's so obvious. But how many people in business tend to do that? Whether it's with things that we're learning for ourselves, do we factor in the time of failure into the cost that we do? Do we factor in the cost of learning how to do something new? Do we factor the fact that we've got staff who may take a little longer to do things because they're learning? How do we do that? And it comes down to that question that we've always asking: are you selling time or are you selling value? If you sell value, you can quote that sort of thing as part of your value, right? But you can't do that if you're selling on time. So it was a really powerful statement that I didn't expect to get from that particular show from a YouTube show. I was like showing about these really cool cars that they were making.

Dave 21:26

Well, and it's I mean, it's a good concept that I think a lot of people overlook. Not just, you know, and and people may call it research and development costs, people call it failure cost, however you want to word it. Like it's it's impactful one way or another. And it does need to be factored into your entire pricing model and everything. You know, and and it was funny, like you mentioned the time investment needed to you know kind of build something. And the beauty of the last I what was I watched, three and a half months now or whatever that Claude's been, you know, kind of really where it's at, but even and it's still improving. And so if you just look, think of where it was three, four months ago to where it is now, it's it's fastly improved. And but for me, you know, the biggest thing I keep saying, you know, I might not be having large in terms of growth or impact in terms of like new finalized products, but I think of all the stuff that we're building for our processes and then clients, and then we're investing that time right now into potential in impacts in the in the near future because we have to make sure it works now to then easily transition it later.

unknown 22:40

Right.

Dave 22:41

So we're working out all the kinks, you know. So some of that's gonna happen in your in your realm when you're when you're building these or you're you're working with somebody to to build it, is you have to realize that it's not just like overnight and then you're gonna have this finalized product, especially if you're building something from the ground up. Like, and we've I think chatted in previous episodes, you know, you have your long term wish fish, but the idea is what's V1, right? What's that first version look like? That it's like, okay, this is Working product, now we're going to like put it into you know effect and then start using it. And then from there we can make changes and improvements once it's viable. And that's kind of what I'm looking at now with a lot of like the command center we're building, you know, for for all of our you know executive and owners is I'm working out all the kinks on my end because I know that when I get those worked out, the implementation side is going to be so much easier, and then it's gonna be that quicker turnaround. So I can't look at the time that I'm doing now and just being like, oh, I didn't really sign a new client today or anything like that. It's be but I did invest right six or seven hours to improve products, and this is when I improved, right? This is the process, and this is the efficiencies that I was able to gain today. And so that's how you kind of have to think of some of these things is that it's not gonna be a direct income right away, you know, but you have to realize what are the hours and are they good hours? Like, you know, did you actually create something or move something forward? And I think that's the biggest piece when you're figuring how do I just get started with AI or how do I catch up? One, you're not you're not that far behind, actually. Like only about 8% of small business owners have really actively implemented you know AI fully, you know, kind of in their workflows and their you know um processes. So you got 92% of people that know about it, right? 84% have played around with it, but only 8% are really using it to actually impact their business on a day-to-day basis in terms of small business owners. So you're not that far behind if you think about it.

Duarne 24:40

No, and one fun fact about that, like I had one business owner tell me today that their friend who they'd referred to me to talk was a huge Claude user, used a lot, you know, was always using Claude, loved it. And so assuming, of course, like, oh okay, cool, must be doing some coding, must be doing some regular usage. I ended up speaking to their referral that they sent through and barely touching the surface, but doing 500 times more than any of their colleagues in the same industry with it. So it just makes you realize everyone's at a different level, everyone's got a different use case. Someone came into my office, was it to yesterday or something? Looked at my screen and said, Why do you have seven screens? That just seems accepted. And I'm like, Well, this one's for coding, this one's for outputs, this one's for monitoring, this one's for chatting, this one's for email, this one's for videos and LinkedIn when I'm not on videos, and like, oh, okay, it makes sense. You've got a reason for having this. Cool, makes sense. All right, good. And but everybody's different, everyone's got a different setup, and we we have to appreciate everyone's at a different level and everyone's trying to get a different output. And like one of the conversations I had with an accounting client of mine back who we supply staffing to, or will be very soon supplying staffing to back in Australia, was around use cases. And one of the use cases I gave was something Dave, you and I have spoken about, where I think the biggest boom in AI for businesses soon is going to be customized applications to solve internal problems. And this is what you're

Building Internal Apps And Productizing

Duarne 26:26

providing to your customers right now. This is something through Triumph you're doing. And when I positioned that, they were like, whoa, I hadn't even considered that. And I said, Oh, absolutely. You know, we've built internal systems. In fact, we set a target for our internal development team that I wanted a plugin for WordPress or an app idea on my desk by Friday to tell me what they suggest would be a good app and give me a use case scenario for that app and who it's for. And I was really quite impressed with the solutions that came through. And the next step of that is go away and build it. You know, you got use a you've got a week to go and build that. Now, that's something that normally wouldn't have been thought of without the use of AI. Without using an AI tool set, you can't really do that. And like I sat with a developer friend of mine, he builds game, he has a game development business. They've got a quite a successful off-road truck game on mobile devices and a monster truck game that's been running for over a decade. And they've done well out of both. And I said to him, I said, let's do some testing with some local GPUs so we can run some local models. So we start testing on his machine, and everything's going well. And he just started, I can see him getting frustrated. He goes, I said, he's like, do this, and it wasn't doing it. It's like take me two minutes to jump in and change the code and do that. And I'm like, yeah, but we're testing a we're testing the functionality on how this works in its native format. He says, Yes, but it's so frustrating to me because this and then he shared with me that the way he works with these tools was very different to the way a vibe coder works with these tools. Where a vibe coder will try and get it to build 99% of the solution by describing where he's actually building over 80% of the solution, then getting vibe coding to help you finish off the more boring stuff or the complex, the more crazy.

Dave 29:20

So he likes the like the detail, the niche of the build, and then he then uses it a different way to then.

Duarne 29:29

Absolutely, and just seeing that was fascinating to me because I was like, wow. But then he was able to help me come up with a workflow for a solution that I want to put in place using localized AI models and frontier models in combination, and it made a lot of sense, and we're able to do things very quickly, very high quality in the testing that we did. And I was genuinely shocked by that. So I think we're dealing with very interesting times for people. There are so many options available for people to use these tools as they go through, and they just need to work out how do they want to use them and what's the outcome that they want to get for their business. Like, what's your goal as codex coders would know? We'll build the we'll build your goal out for you.

Dave 30:23

And and I I think you know, what when you when you get dipped down to it, I don't think like you have to be where like we're at, right? I didn't start here. You know, I think you and I, in terms of AI, have been adopters of it since since the beginning. And while it definitely was not, it was more frustrating probably at the beginning than it is now for sure, with the inabilities that it had. But but I think what it really comes down to is you know, just start. Like you're not never like everybody starts from nothing and then you grow. Like AI is the same way. You're gonna start, you're gonna get an understanding, you're gonna say, Hey, could I how how can I do this? And that's what you need to be, you know, when I think about it, is you have to be curious, like be the curious George of the situation and just chat and just say, you know, I've had this idea. How would you approach this? And and that's a beautiful way to, and then you're not trying to, it's gonna give you the steps that it would take right and say and walk you through it, you know, and and give you that insight that instead of saying, Hey, I want you to go build this, this is how I want you to do it, where maybe you're newer and you don't know how to do it, just say, you know what, we have this problem. How can we potentially solve this? What would you do to solve this?

Duarne 31:48

That's it. Research this time, we find out how other people, yeah.

Dave 31:54

And so, you know, when I think about it, yeah, I'd love I love helping people build something. But here's the thing, here's the thing, and I've said this before, like, I'm not doing anything that you can't do yourself. I'm not. Like everybody can do what we do right now because of where we're at. So I'm not selling you some secret sauce or some different thing. Uh, what I what I am and I talk to my clients about, or or what anybody's doing is is they're you're partnering with them because of the trust you have in in that person to help you get the result that you want. One. But two, like they're saving you time, they're saving you energy where you don't have to you know do it yourself, which is the biggest thing I tell a lot of people. It's like, hey, if you don't have 13 hours a day to like keep up on it and maintain it, like, yeah, I get it. Like, that's probably why it would be a good partnership here. The anxiety around it could be something that is frustrating or people don't want to deal with, is another one. And then that partnership is a big one. It just the advisory, the the outside thought, the you know, hey, where are you at? What's going on? Maybe a lot of mechanisms, you know.

Duarne 32:58

The one I'm speaking at right now is people are ex they're able to explain what they want, they're able to get to a point, and then they get to that point where, like, like if you're building an app, for example, you have to have an understanding on where to host the app. You have to be able to do something with so you can get the app to display in a format that's usable, right? And they can explain that, but then they get to that point where go, well, I don't know how to launch it, I don't know how to set up a hosting environment for it. And yes, these tools can help you with that now, but having a basic understanding and technical background does certainly help with that. And for some people, they just go, you know what, this is a little too hard. I'm gonna throw in the towel, I don't want to do any of this anymore. And that's where I totally get utilizing outside sources and getting them to help you, whether it's train you or do it for you. But one thing I pointed out to a client the other day was you can kind of if you need custom internal software nowadays, you don't have to go spend $100,000 on getting it built and take six months to do it. It could be done in a couple of months. And if you think about it, like we have an internal project at the moment for a client, and it was presented to me today to review, or actually a few days ago, I got a chance to review it today. The plan, and I'm looking at it going, well, this is great, but have you considered that this is something that other people in that our clients' industry is going to need? And they may not be using WordPress. So maybe instead of building it as a plugin in WordPress, why don't we build this as a script that can be ran over the top of any website and as an overlay and do what it needs to do? And while we're doing it, why don't we brand it as an internal product that we can then resell to other clients in the future? So we'll productize it. And it was suddenly just a mindset shift. And it was like, well, yeah, the client could pay for that, or we could get the client to invest in their own solution, we'll customize and build it for them, we'll subsidize a portion of it and build a solution that we can resell in the future as a product to others. That's not completely narrowed in to solve one problem, it solves one problem across multiple client bases.

Dave 35:21

Well, and that's kind of like you mentioned a lot of people like they probably find a solution or software, and they're like, Man, I love this, but it doesn't do this. Great, like you could build something similar, but that does that, you know. Or, you know, if you don't want to take the time, then you just have to deal with the missing link that a lot of people, you know, sometimes don't want to. And and you know, you kind of mentioned I think the the one thing you have to consider, which a lot of people don't, they're like, oh, I'll just build it myself. You know, there are going to be ongoing costs that you still have to pay, right? Especially as it becomes a larger platform database, as you have you have storage costs, right? You have you know processing costs, different things that you still have in the background that yeah are manageable. But if you're talking tens of transactions or processes a week versus tens of thousands of transactions a week, the costs are gonna be different. So there are some things to consider when evaluating, hey, what what do I want to build? And does it make sense for us to custom versus not? Absolutely.

Duarne 36:27

And the other thing you got to factor into is integration into third-party platforms. Whether it's an API token usage, there's a there's an ongoing cost there. If you build it, is it going to support multiple platforms? You can move that in the future, or is it locked into one? Are you hard coding? Are you making it a variable? Are you and every time there's a change in a third party, like if you're using something that needs to tie into the API of Xero, so that's the back-end functionality of Zero to get functions, or the back end of QuickBooks, you need to factor in that they change those time to time, and you'll need to rebuild that portion that works to make sure it talks properly. So it's not a set and forget. There is going to be an ongoing maintenance, there's going to be an ongoing uh ability to fix and check what's going on. It's not a one-time cost set, you know, one and done. And that's something that people generally overlook. I like I was saying to my brother earlier today, and he was saying, Oh, look, I'm loving Claude. I'm doing some great things with Claude, I'm building these apps, I'm going to be doing this, you know, this idea that I've got. And I'm like, that's great. Have you put any work into hardening your apps yet? Question mark comes back. What do you mean, harden? And I'm like, Well, you've got to build the security and then from you know, make sure your compliance is in place and make sure the integrity of the app is there so it can't be, you know, your customer information or your sensitive information inside the app isn't compromised. And it just opens up the doors of a little education when you're doing this helps. It's not it, but getting started, you don't need to know that right away. This is where just educating yourself along the way. And like I said to him, I said, don't stress too much. Here's a prompt to give to Claude when you're ready that'll help you build out and it'll and it'll ask it, what can I do to make this better? And he was like, Oh, that's cool.

Dave 38:29

Well, I mean, I can't remember. I wish I would remember the guy's name. But he's like a coding guy, and he had, you know, and the beauty of when you're if you are getting started on some of this stuff too is you know, GitHub, which is obviously like this open, you know, one, if you have coding, you're gonna store it there. But you can choose private or public, and a lot of these sort of like companies or these bus these private people too, they will post a public GitHub and that shares some of the skills or some of the different things that they're doing that they are sharing with their team. And on their YouTube videos, they they share the link to their GitHub. Now, obviously, if it's something that maybe has two stars versus 90,000 stars, you might want to trust the 90,000 stars and stars as a way that you can like upvote things on GitHub, right, that are open source. But you know, I found one the other day for coding, and it I've used it now a lot. I use it a ton, and it's a command, it's like in it's um improve code archis architecture. And the the the point is that it will literally you can point it at the entire app, or you can point it at one singular piece and and what you're trying to improve, and what it'll do is it's gonna analyze it, and then it gives you an actual report and says, Hey, here's here's what your current version, here's how I we suggest that like you would improve it. Like, you know, maybe you have like three different pages that are individually calling an LLM where the improvement would be have all three go to one caller so that it's consistent instead of it being you know three different versions, and then it will give you anywhere from two to six suggestions of how to improve. And then once you go from there, it will then walk you through an interview process. Hey, here we're gonna go through number one like is this what you're trying to do? You know, does this make sense? It'll keep analyzing, it's not gonna change anything yet until you actually then approve it afterwards. So there it's it's a really cool skill that I have now been using. I pretty much use it five, six times a day. But ultimately, the goal would be on every platform is that you, as you're making changes, you want to make sure that what it's doing is still not gonna break anything. So you just run it as a you run it as a routine on every one of your projects at the end of every week. Hey, you know, run this skill and then just review the changes from the week and make sure, you know, are there any suggested improvements? And it says, like, hey, this is mandatory or this is like you worth research, you know, and then it gives you that sort of response so that you can decide what you want to do. So that's been something cool that I found this week. And I'll have to find uh the actual YouTube video from the one that I found it from, and the gentleman that that you know I watched the video on. But so there's things like that where find resources, find different things that you know make sense, and then follow them, but then pick pick and choose what you want to implement. Like you don't have to do everything, you don't have to match somebody one for one. There's a lot of people I subscribe to that I don't watch all their videos because what they're talking about is not something I'm trying to do right now.

Duarne 41:30

There's a lot of hype out there, too. A lot of trends, a lot of hype, a lot of crap just to try and get the followers and get the because you got to remember every time you get a play on a video, that goes towards getting uh getting paid. So there's a lot of in sponsorship deals and things like that.

Dave 41:46

So and what I've noticed too, and I don't know, maybe warrant, hear me your thoughts on this one, but what I what I have noticed was one,

Cutting Through AI Hype Safely

Dave 41:51

you you you hit it, right? They're though they're content creators probably first, AI second, or maybe they were AI and then they're now their content creators. But their goal is to talk about every single thing that's changing, every single thing that's coming out. But there's no practical implication of what they're doing, right? There's no like, hey, this is how you should be doing this if you are X, Y, and Z. There's none of that. And so that's why it really is important to take a lot of that high-level you know, hype, as you called it, right, with a grain of salt and just figure out what makes sense.

Duarne 42:27

Well, I learned this a long time ago during crypto, right? I mean, crypto when I was doing my crypto mining, crypto trading four or five years ago, there was so much bloody crap coming out on such a quick pace on YouTube, you couldn't keep up with it. And I realized if it was already on YouTube, it was already old news. And it was it took so long to get up. If you're watching anything and it's AI related about Claude, Codecs, Gemini, basically any AI coding language, and it's more than seven days old, there's a bloody good chance it's outdated. And that's the truth for a lot of that stuff. Now it does it's not a general how-to guide, like in the traditional sense, like how to set up a piece of software, because the stuff is changing so rapidly. But in the same regard, don't find just be careful of that hype because my phone, I'd subscribe to a bunch of channels, and you'll get the long form converted into the short form, and then your phone will just blow up with all of these little short forms for long form stuff that you've already watched, yeah, and give you the snippets. And sometimes you get some good stuff out of it, but a lot of times it just it's a lot of noise at you. And for me, I found a a lot of the ones I just started unsubscribing to were oh, the new open claw just came out. You need to know about these features. Oh, they've just changed the way open claw is gonna handle the you know SEO. That's like it's too much.

Dave 44:00

And you said it, right? They're you don't want to call it sensationalizing, you don't want to call it clickbait, right? But it kind of is because they have to get you to click on the video and watch. Yes, there's been a lot of times lately, which is why I actually now that I'm in it and just doing it 100%, I don't watch as much. But when I first getting started like four months ago, like I was trying to consume anything that I could. But what I started to realize was that they would come out with a new video or a new topic or a new you know headline or whatever, but yet the majority of the concept in there was the same of stuff that they've already said. There was nothing really new that they were talking about, they were just releasing another video to you know in a different way to get the content out again. And that's that's their goal. And if you realize, like that's their goal, they have to put out more videos to get more views to get paid. Whereas somebody like you and I, we we're like, no, well, fuck the views. Like, I don't know. I I mean, I love doing this, like, and I love the that you guys are watching this, right? And I would be happy if this was you know continues to grow, which it is. But our ultimate goal is not necessarily this. Our ultimate goal is how do we continue to implement and solve people's problems? And yeah, I I know for a fact if I if I wanted to, you know, we could start putting out videos like that that are specific and how-to's and all of that, but nobody will one. I just don't have the time to do it, but we do all that kind of in a you know our our Triumph Studio where we help people kind of implement that stuff. But I just don't have the time to create the videos, but that would get tons of stuff if you just did that.

Duarne 45:37

Well, that's right. And you gotta remember, these a lot of these content creators are actually paid to create content. That's their business, that's how they make money. So they'll sit there and create all this content. A lot of them, not all of them, and that's how they get paid. They get paid. If you look at them, like oh yeah, I got sponsored to by this company, and they've just sent me four Mac Studios worth ten thousand dollars each so that I could do this test and show you how well Kimmy runs across all of these platforms as an AI, blah blah blah. Watches I fail 50 times and eventually get it right. It's great and it's fun to watch. And then you realize that it's a sponsored by the couple of companies, they're open about that, but you've got to take it on face value.

Dave 46:28

But if you're watching a lot of Apple Apple send them to studio, like all four of them, like again, like you listen, like they say it. They're like, Oh, yeah, you know, by the way, like just letting you know, like I was sent these by Apple, like I didn't pay for these.

Duarne 46:39

You know, I would never pay for these, these are so expensive. I've got to send them back in a month.

Dave 46:44

But the ultimate, the all it's all content, you know what I mean? Exactly. Ultimately, it's from the people sponsoring the videos that are trying to generate sales because what will happen is people will see that they're gonna be like, That's the best thing, some sliced bread, not thinking how do I practically use it, right? Yeah, and then they they go out and they buy these Mac Studios, maybe not four or five of them, but maybe they buy two and they don't need them, and then they end up wasting the money, and they just don't ever practically implement. And that's where I saw a thing the other day, and it said like before AI, and then it had like two, you know, the typical like people waiting behind a door, right? But the two things were I think it was idea and then execution. And the one thing I didn't disagree with is you know, those were kind of the same size of group where the idea group and the execution group was roughly the same. But the thing I didn't I disagree with is that I still think even before AI, there was tons of people with ideas that they just didn't know how to execute. So, like if I were to do that same diagram again, it would have idea group still huge, and then execution group you know, smaller, right? Smaller line. And then said after AI, and and how they changed it was there's even more people in the idea group, and then even less people in execution. And I kind of get it because if you think about it, there's now when you think of AI, there's just so many more it just opens up many more lanes. And so while I do believe there are way more people in the idea lane, and things that they have an idea, maybe they start to build, but they're never gonna never gonna finish. I I do agree that on the actual execution, the actual deliverable, there's gonna be less people. And that's where it really kind of comes down to you know, uh, does it make sense? In the end, whatever you're thinking or wanting to do or build, does it solve a problem? Does it save time? Those are the two things that are gonna determine the success of a platform or a dashboard or an internal workflow, anything. That those are the two things that are going to determine if it's gonna be a success. If it's just uh a knife to have where it adds more cost, or you know, it wasn't a problem to begin with, but now it's creating problems. Well, that thing's gonna go by the wayside pretty quick, you know what I mean? So you have to do a lot of analysis on your processes first before you even get started in AI.

SPEAKER_01 49:27

Oh, yeah.

Duarne 49:28

You gotta do the research. I mean, it's a bit like you've got to know what your outcome is at the end. How many times are you know those um car shows? I love watching car restoration shows, right?

Dave 49:39

And I couldn't tell from your earlier message.

Duarne 49:43

So you know when you watch a car restoration, right, and you're watching if you've got if they've got a client in mind, like Counts Customs, for example, have a client in mind, they build to suit the client, right? But then you've got these other guys who they get on there and they build a car so sensational and crazy the way they want it, that then they sit there and struggle to sell it for months and months and months and months and months. Now, I totally get that because we used to do the custom bike building, and we would build a custom bike, and then we'd paint it the colors that I thought were pretty cool bright, outrageous colors, and we'd do some fun stuff with the paint and the design and the styling, and it was great. A lot of people appreciated it, but not a lot of people were willing to shell out the money because they just wanted a black motorcycle. That's it. They wanted the cool things that we did, but in black. And we actually had a bunch of people, hey, well, if you repaint a black, I'll buy it from you. And I'm like, nah, I'm good. And that's the same sort of scenario, right? If you want to build something, you need to understand who it's for if you plan to sell it. If you plan and if you're building internal software, it's really easy because you know who you're building it for. It's for internal use, you know what your case is. So if you are going to start building something, I would recommend don't build something to sell for your first project. Build something to solve a problem you've got internally that you understand intimately, and then just improve on that. And once you've done it, you'll find yourself going, Oh, I could solve this other person's problem with this product as well. And then look at building that product out, but do it from the point of view of what problems are you solving with that product you're building.

Dave 51:28

And when you got to realize too, like most the majority of you here that are watching the show or watching replay, whatever it may be, you know, you're you don't need to focus on anything else besides internal. Like if your business is not focused on providing support in this arena, like you shouldn't be thinking outside the box right now. Like your whole focus should be internal.

Two Lists For Your AI Roadmap

Dave 51:50

How do I become more efficient? How do our teams become more efficient to do more with less and free up more time to do what makes us money, right? And you know, when you think about that, and it kind of goes back to like two lists, and I've I've talked about these two lists in previous episodes, but I'm gonna bring it up here again because it it's where you have to get started, and I talked about it in our panel discussion yesterday. The first list, and you have to do this before you kind of determine where you're gonna go, but the first list is what are all the things that you are doing right now manually that are taking up your time? And you would like to have some sort of automation or it be fully automated because remember, some some workflows make sense where it's you know 80% auto or AI generated, and then there's a you know 10% at the beginning or 10% at the end, you know, human element that just reviews and approves. That's your first list. So that's all of those things that you're doing. That's taking up all your time. Write down how many hours it's taking you that you're just like, man, I wish we didn't have to do this anymore manually. I wish it just kind of gave us the output. Great. That's list number one. List number two are those things that you know you need to be doing or that you wish you had more time to do. And this could be internally, it could be externally, in terms of, hey, I wish I had more time to spend time with my family, or I wish I had more time to, you know, put out more social media or create more videos, like whatever it is, put that on list two. And then those two lists, when you're done, are going to kind of point you in the right direction of where you should be either creating something or developing a workflow, et cetera. Because some of your stuff on list two are just gonna be like, oh, they we can develop a workflow for that. And then we give it the information. Most of the stuff on list one are gonna be the things that you're gonna eventually build a workflow on. So they're going to direct you and generate your roadmap. And the next thing is don't try to do it all at once. Start with two or three, and then get those completely done and then work through it. So prioritize your list because the biggest thing I hear from a lot of people is they want to solve everything at once. They're like, I have all these problems and I want AI to solve them. Okay, great. Which one do you want to start with? All of them. I'm like, no, you're like we can't do that. Your attention is gonna be diverted, and plus, do you have the 40 hours a week to develop to do all of them? Well, no, no, I don't have all that. Like I still got to service clients and get you exactly. So, what's the one thing right now that's going to give you the most impact? And then let's start there. And that's what you have to focus on in your business or uh internally in your organization, if you're a bigger organization, that's where you put your attention first, and then you work through it. And the one thing you've mentioned, and I'll let you go back to it, but make sure you have the stakeholders involved in the process as the owner, CEO, whatever you are of your business, or you know, maybe your C level. Make sure when you're working through an AI process that you involve every level that is gonna be impacted by that AI.

Stakeholders And Process Mapping

Dave 54:52

You know, so it's not just like, oh, we're gonna go do this at the C suite and the executive level, and then we're gonna force it down to everybody else. And then once it's kind of built and developed, then we'll give it to everybody. You're you're gonna miss things that are overlooked. And trust me, I get this from my my executive experience in corporate. You know, there's a lot of things that we think are happening in the processes that once you actually have that conversation with the team members, you realize you know, maybe steps three and four were like dropped years ago because they they weren't efficient anymore or they weren't needed and they just did it all in step six. You know, so but if you build an AI to do three and four, like now you just wasted time. So it's so important to think of all the stakeholders throughout that process and involve them all when you're going through it, especially if you're you know a bigger organization, even if you're 10 employees, that that's so important.

Duarne 55:42

If you're an individual employee or three employees, right? I mean, what like we witnessed earlier. I mean, just because you have systems, if you don't actually review your systems, if you don't check your systems, you don't know what's happening, and that can be a huge detrimental cost to you and your business in the future. One of the things that we witness is there's assumptions made on how people do tasks in our business. I've had this so many times, where I'm just assuming something's done one way, then I'll complain to the team why is this taking so long? Or why did you know why am I getting this result? I thought we did it this way. Isn't this how we trained it? Oh no, we've been doing it this way because that was what we thought was the better way to do it. And it's like, well, hold on. I designed it specifically this way because that needs to be done so I can do this, this, and this. Because I understood the whole picture and I'd never shared the whole picture because at the time I didn't think it was a value to do that with the staff because they didn't need to see the whole picture. Well, AI is different, AI needs to know the whole picture because if it doesn't know the whole picture, it can't warn you about things that are coming up that you need to be considerate of and aware of.

Dave 56:56

Um it's going to, you know, it will make up the steps if it needs to, right? And that's what you don't want to happen. So that's why you want to have the entire process mapped out first, and then you give that instruction to the AI process to work through it. You know, create it into a document, a workflow document, and then at every level, right? The most important thing to do would be each level has its skill, right? So that it's repeatable and it goes back and does the same thing over and over again instead of just saying, hey, go run through this process. Like you have to create the skills associated with it. And so that's why it's not as simple as just like, oh, I want you to create this workflow for me. Okay, what happens from there? And I'll give an example this past week. One of my clients is using Claude and connecting it to Go High Level.

Why AI Acts Like An Intern

Dave 57:51

And so part of the process was hey, go to go to Go High Level and build out the workflows, build out the automations and the landing pages, and let me know when it's done. And I think Claude's like, okay, do you want me to go do it now? And the client was like, Yeah, yeah, go do it. Reports back that it's done. And now it created these things, it created different workflows and created the different steps with the different copy and all this kind of stuff. But the client didn't know what to check for. The client didn't know because the client's never done any of that stuff in go high level. So then when I get in and I'm like, okay, let's take a look at what it built, I realize that how it built it was inefficient. You know, all the workflows were being kicked off by by tags being added, but then there was no automation to add tags to the prospects. So it's like all the tags that had to be added on you know manually. So it's like this isn't an automation process, it was expecting people to book without booking through the appointment schedule, which is typically what you would want to have your automation trigger would be hey, you somebody booked into this webinar, into this you know, work set worse work session. So then put them into this automation. Well, no, it's like, hey, when you add the tag work session registered, now we're just gonna go into this automation. So this is where you know AI is good at developing process, but you have to remember it's like an entry-level intern. You have to be able to give it that direction or have trained it in a specific way to do those things over and over and over again. And if you have it, it's it's gonna give you that basic level intern output. And if you for that same process are an intern yourself, you're not gonna be able to check that it did it right. So part of the result for that was like, yeah, it just it didn't do anything. It like I thought it was sending out messages and and getting people registered, and and then when I came to the registration time, like I realized like nobody was actually registered. And I'm like, Oh, did you double check? Like, did we what what's going on? That's how we kind of jumped in and kind of looked at it and realized like none of this is correct, you know what I mean, in terms of like how it set it up. 90 percent well, I'd say 80% of it was set up, right? But it's at 20%.

Duarne 1:00:13

You know what it comes down to still what this is this is something that I was given as an analogy recently that I heard, and I thought it was pretty spot on. When you build a house, do you go to the builder or the general contractor and say, build me a house? And then just let them go hard, or do you go and talk to an architect first? You get your architect to create the design, you work through the architect, you make sure it's right, then you go and work with the you know the engineering to make sure it's actually going to do what it needs to do when it's bearing load bearings are right and stuff. I'm not a builder, so get me dumb. I may be using the wrong terminology here. You probably need to consult your interior design to make sure you've got all of your materials inside sorted out correctly. And then once you've got all that sort of cool stuff done, you then you go and include all your trades. And there was a scenario recently explained to me that somebody went and built a bunch of extensions in this building without consulting the data cabler, without consulting the electrician, without consulting the plumber. So none of this was installed in the building prior because basically they had a the they didn't get an architect and an engineer in first to take a look at it. It was just an extension. We don't need that. And then when they come in after, it's like, well, we've got to dribble the we've got to rip all that dry drywall out, we've got to be able to get access to that. Oh, that's gonna, we're gonna have to get all that back put back in. So, well, yeah, you're gonna get your you know, you're gonna have to get your carpenters back in to do that.

Dave 1:01:52

And so this is why, too, you know, it I don't think like AI is ever going to take over everything from this piece because of the fact that there's still that knowledge of the individual workflows, processes, whatever you want it to do that a human needs to be involved with. Now, maybe in in the future, will it have all knowing knowledge of everything? Maybe, but just think of what that means, just think of that backend that they would need to have all knowing knowledge of every single software platform that's ever been released and how it should be done. No, yeah, that's why you're unique as a business owner with your experience. That's why your AI is trained differently than maybe Dwarne and I's AI, because you're unique in what you want it to look at, right? When you think of a go-high level, and I'm sure there's people out there doing it right now, they're probably training some sort of AI process on how to do the workflows efficiently, on how to build them correctly with their skills. Yep. And here's the beauty of some of that. We mentioned before there's probably some public repos out there. I'm sure there's probably a couple public repos out there that have open source skills related to Go High Level or any other platform that you can copy the link to the GitHub repo. You go to your claude and you say, Hey, I want to potentially use some of the information in this GitHub repo. Analyze the repo and tell me what we should be doing and what we should bring in or what we should create.

Duarne 1:03:36

Analyze first, don't just make something. It's the same thing. If you're gonna use like a tool, make sure it's the right one. Like, if you're

Open Source Risks And Security Checks

Duarne 1:03:56

gonna integrate Stripe into your website, you probably want to use the Stripe official supported plugin to do that, not some third party that's done it, because even though the third party could do it, it might have a bunch of security vulnerabilities that you know you're not aware of. You're not a developer, you're not gonna know that. And it's the same thing that goes for these skills. There was actually a big problem that I was reading about that when OpenClaw first came out, there were so many skills and options and stuff uploaded in such a record time that people were grabbing and using only to find out that there was malicious code in them. So they had malicious code in there that was potentially opening up vulnerabilities on their computers and people and keylogging and all sorts of crazy stuff, and there was stuff they didn't even know about, and it was some of it was malicious, some of it was just unintentional bad coding. It was bad information, right? Because it wasn't double-checked, it wasn't triple checked, and it all comes down to, like you said before, if you don't know, use the tools to check it for you. I had an email XML file sent through the other day from Google, it looked official, and I dropped it in chat 2D and said, I want you to analyze this and tell me, is this an official Google XML? And it went and checked and he said, actually, yeah, it is. Based on where it's coming from, based on the information provided. And the good news is it's all tech, it's all good, it's all ticked. Great. I'm not an expert at reading those files, and I didn't want to open it on my computer in case there was some problem. It was able to open it and view it without opening it on my computer and do it safely. And I let it do it. It took the risk for me and it gave me the answer. I got a professional answer without being a professional. And I guess that's where the scenario is. I've been building computers recently for uh projects I'm working on, and I get stuck sometimes. I'll take a photo of the motherboard and go, this memory configuration is not booting. Can you give me an indication? Oh, you've got that in the wrong dim slot. Move it on this particular model, you have to put it on this bank because that bank doesn't work unless you're using this type of CPU. Oh, of course I'm gonna know that. I could go and look that stuff up and watch a hundred videos and read Reddit, forums, channels. Eventually I'll get it. Or I could just keep trying. Or I can ask the question and get it to help me find an answer quickly, or at least point me in the right direction to test new things. And that's where you can find so many use cases with AI, it's not even funny anymore. It's to the point where I ask myself, am I going to go on YouTube and try and find a video on how to do something? Or am I just gonna ask ChatGPT in my instance to go and search for it myself and say, hey, go and look on YouTube and see what you can find about this while you're at it, go check Reddit and see if anyone's talking about this and make reference to any videos that I should be watching about this.

Dave 1:06:59

Well, and so and and so those are those are the cool things that, and then we'll drop it with this because we're already over an hour with crazy how quick time fashion flies when you're talking AI. But you know, that's where creating your repeatable process, like if you want it like for yourself, right? You have your research that you'd like to do, building that research into an engine or a repeatable system on your claw or wherever you're using, whether it's you know, code, and this is where I've I've really moved a lot of my processes off of just clawed code co work and actually move them into creating specific engines in code because I don't know if you've you realize this, but like claude code actually can now do a lot of the stuff that co work was doing in terms of like going to the browser and doing all this different stuff that. Cowork kind of hung its hat on, which I get it. Like, and there's still some things that I still have to use cowork for that code can't do. But code, you can get a lot more in-depth, quicker, quicker response times with the CLI and all that stuff instead of having to go out to the website, pull it down, open it up, all that kind of stuff. So there are some efficiencies there. And so taking like a research engine as an example and telling when you go through the first process and you've gotten the result that you wanted, okay, great. I want you to take what we just did and I want you to turn that into a skill, a research skill. And then from there, going to your code and saying, around this skill, I want to build a research engine where I can drop an idea. It pulls ideas for me at the beginning of every day, but I can also give it inspiration to go find something if I needed to. And now you have a repeatable system that you don't have to like go and think of how did I prompt that last time to get the result? That's the beautiful thing right now is that you don't have to continually remember. You're not, you don't have to be a prompt engineer anymore to make things impactful. The idea now is how do you just build that repeatable system to get the same results? So I'll leave it with that.

Duarne 1:09:03

And speaking of clever, right? Using these tools that we've got and using them in different ways, like Mike, I'm using Codex this last couple of weeks, building out some apps, and I've got it automatically set up to my VM via SSH for access with its own unique login details. And I say, Hey, I want you to go and check my VM for why this isn't working. And it goes over and it fixes the problem. I want you to report to me but and ask for approval before making any changes. Never make changes to the data. You know, so and it comes back to all right, did not change the data. I need to do this, I need to restart this service. Is that okay? Sure, go ahead. Like you can use these tools to help you and guide you in directions. Knowing what it's doing is really helpful because you want to make sure you're doing it right, but get those mistakes, get make a mistake. And I had one instance where I overwrote all my data, my test data that was in the system, and I had to go and put it all back in. Look, I just happen to have enough exports of certain information to be able to get codex to put most of it back in for me, but I still had to go through and fix a lot of it. So these systems are smart, but they can also, they're not foolproof. So before you just go saying yes, yes, yes, yes, you need to understand some basics with it. So I really love the idea of building internal apps, though. I really love the idea of just taking something like as simple as a spreadsheet and then utilizing that into a dashboard that you can use internally, and maybe even taking an old access database. I know I'm going back a bit there, right? And converting that access database into an interactive

Dashboards From Spreadsheets And Databases

Duarne 1:10:47

app that you can use.

Dave 1:10:49

And that's the beauty of, and and these are and and hopefully, I mean, my takeaway that I hope you guys got from today's episode if you're watching this or you're watching a replay live. Taking the idea, like the things that we're saying here, practicality-wise, and think of how you can do those in your business. You know, like like taking something that is supposed to be a tracking spreadsheet that you, you know, everybody has access to and then they post all their information to regularly, like putting that into a quick online KPI dashboard that works on mobile, that instead of people having to log in and go to the spreadsheet and find it and implement the data, they could just open their phone and drop in the data right there from the results. Like, how do you make your people's jobs easier as well, your employees' jobs easier? Like, that's where this all comes down to. And you mentioned it. You know, you may think, oh man, it's gonna take a long time for us to like take an old access database or a spreadsheet and get the information in there. Here's how freaking awesome like this stuff is is you could take that database, give it to your clawed code, and say, Hey, I want to take all of this information and then I need to backfill the data, and it'll do it for you, you know, and it'll fill in whatever, you know, and so you got to figure out yourself what database you want to use, whether it's Firestore or you know, a super base, wherever your back end build is going to be.

Duarne 1:12:11

But now you have it in a virtual environment protected, you know, it's not on a spreadsheet on your on your computer, and it's backed up, you know, all of that kind of stuff, so that you're able to set it up to be accessible from anywhere in the world if you set it up correctly. So one of those days where you have to call Melinda in the office because she's the only one who acts has the access to the computer sitting in the back room behind your deck behind your desk where the access is sitting on that computer that's 15-20 years old because you can't throw it out because it's running the perfect software of that version of the database.

Dave 1:12:52

And that's where like you know, it's one of those things where the biggest thing when I was in the corporate world was that we want every one of our managers to have their own dashboard, they can see their own information. Well, you know how easy that is to build now, but we went through four or five different, you know, sort of accounting CRM softwares back then, you know, and they wanted what were they charging? I think the the one we ended up on, it's like $35,000 a year for access for all this information. And I'm just thinking to myself, holy shit, like just how easy I can create that now, and and build it so that it works with the authentication of who's logging in and what their roles are. And man, if I if we had this access back when I back then, like we we could be so much more efficient. But the beauty of that now is that every one of my clients one first thing, you can do it yourself. Like, you don't need me if you want to take the time and do it yourself. Like, you don't need Dwarne and I, you know, in your business to do it. Be honest with you with that. If somebody says, Oh, yeah, you can't do this, they're lying to you. Anything that we tell you, you can go and do yourself. Or this is why we're saying it the way that we are, you can do any of this. Right. The biggest thing that things like in and businesses like Dwarne and I are is that it saves you time, it saves you the anxiety, it saves you the stress, right? Around it. And some of the time, you know, for me, I set it up and we handle a lot of the back end in terms of like the cost and stuff like that, so that you don't have to just built in so that you're not trying to constantly update it and worry about it. So those are the things that you have to really analyze and figure out you know, do you want to invest all the time into figuring out how to build it, or is it better to just work with somebody and partner with somebody that you trust? It doesn't have to be us, but find somebody you locally or anybody that you trust you're like, hey, let's do this, you know. And that's where I want you know, want you to kind of take away is there are you can do anything we talk about. Anything you see right now, the the ability is there for you to do it all yourself. You just have to decide do you want to do it yourself or do you want to partner somebody to do it?

Duarne 1:14:52

It's an upskill. And if you're not willing to upskill, or you don't have the time to upskill, or don't have the capacity, some people just mentally don't have the capacity to upskill. Or it's not the right time to do that.

Dave 1:15:04

You have to have a passion around it. Like I said that you know yesterday in the panel was you know, how do you how did you get where you are? And I'm like, Well, I'm passionate about what I was what I'm doing right now. Like, I'm I I come into the office and I could be here for 13 hours a day, and it does not feel like 13 hours. The time flies, but I love what I'm doing. Yeah, and you know, my I think sometimes it's

Passion Priorities And Closing

Dave 1:15:26

hard for for the my fiance to be like, Well, how are you still at the office? I'm like, because I'm I got so much to do, I feel like I'm behind, but I love what I'm doing. Like, and I want to be home, I get it. But there's this is that the moment of being a business owner where it's hey, I'm putting all this time in for this like year, two, three years to get it up and running, to then the next 15 afterwards, I don't have to. That's the ultimate goal, right? That and so that's why you have to be willing to kind of do that and have that passion. And if you don't have the passion around it, you're never, you know, you're not gonna do it. You're not gonna wake up at 7 a.m. and then be here, you know, be at the office until 8 p.m. You know, you're not gonna, you know, be willing to sit at a desk for that long. You're not gonna want to work on Saturdays because you want to continue to build for people, you're not gonna set up client meetings at like Joe Dwarren, it's you know, 10 p.m. at night for you. You're not gonna do that if you're not passionate about what you're working on. But you have to be passionate in order to make it a success. And if you're not passionate about AI development, then don't try to be. Just because you can do it. That's the that's the thing you have to decide on. That's the biggest piece.

Duarne 1:16:31

Same thing goes. If you're already, say you're a trades person and that's your business, and you're already on the tools from seven, and you knock off at seven, and you're already dealing, and that last couple of hours is dealing with the accounting part or dealing with your marketing part. The idea of doing something with AI and learning a whole new skill may just be the most insane idea for you, or the most over-the-top time for you to be able to do, and that's fine. No one's saying you have to, or it's but you might want to find a partner who's willing to invest the time to do it for you, then you can get some of that time back. Maybe that extra couple of hours of an evening you're spending doing some of that marketing or doing some of that accounting could be bought back into as an automation or some AI tools that you use in-house. Another really big one is that the amount of industries that have, like you mentioned before, industry-specific software that charges a fortune to run. And some of this stuff is just insane. And it's thousands of dollars a month to run this stuff. And in a lot of cases, you could build out a lot of this stuff to solve your problems because that software might solve 50, 100 problems that you don't have. But it solves one of your problems. So if you can solve your one problem without having to spend those thousands of dollars, then maybe building something or having it built for you is a really good solution because you're gonna save money long term. It's a bit like installing solar on your roof, right? You're not gonna you're gonna spend a lot of money up front, but you're gonna get savings ongoing from your power usage.

Dave 1:18:16

Well, and this is where you know, and I know we're we're way too long right now. So we'll we'll wrap it up now. But the the ultimate thing is is you just have to realize you you said it, and this is what we said, I said it earlier. What is the priority to save you the time now to free up you, the business owner, and your team to do the things that are actually going to drive the business forward and grow, right? If you're not able to do more estimates or you're not able to you know generate more interest or more leads because you're doing a lot of these things manually, that's what's hindering your business. And what are those things that are holding you back? Those are the things that you need to look at and and build. And if you're not willing to do that, then I guess you're not really dedicated to your growth.

Duarne 1:19:07

And that's where you may be in the wrong position, right? I mean, and you need to have a reass you need to do that reassess, re-evaluation in with yourself and your business. We've talked about in another episode, like have that hard look, have that hard conversation.

Dave 1:19:24

Yeah, but well, with that one, what do you hope? What's your thing that you hope somebody takes away from today's episode? I already said mine.

Duarne 1:19:33

Look, I hope that people stop paying attention to all the hype around AI for a minute and actually give it a go and try solving a problem that you've got, or at least asking questions to learn how to use the tools or achieve an outcome that you're trying to get, yeah, and then decide if that's something you want to proceed with with these tools that we have available. You'd be very surprised what you can achieve with a $20 subscription to Chat GPT. You'd be very surprised what you can achieve with a $20 subscription to Claude. Um, it's very intriguing what they actually offer at the moment. Take advantage of it because these companies are getting a little bit more stringent on how many tokens they're gonna give you in their packages. It's kind of like getting you addicted to using the tools right now, and then eventually the prices are gonna go up or the usage requirements, usage capabilities are gonna go down. Right. So take advantage, get in and use it now, solve some problems now, and work out how it's gonna work in your business. And just one thing to be really careful of is if you are getting started, avoid the API tokens for now. Just focus on the or you can eat buffet where you're signing up for the $20, you can eat, you're still limited with the it's like there is rules and limitations to this. So the $20, the $100, the $200 plant. Just start with those. Just I I spoke to someone the other day who was really proud about building a website and he spent $400 in tokens. And I looked at the website and I was like, oh, could have done that with a $20 subscription. It was just approached it the wrong way. So, but yeah, be willing to fail and learn from your mistakes and have some fun while you're doing it, and hopefully you get as passionate as we do.

Dave 1:21:34

Or and it, but if you don't, that's okay. Like you probably have a passion somewhere else in doing what you're doing in your business and serving your clients, and that's completely fine as well. So, whatever works for you. But with that, we appreciate you. We love you. We we hope you guys are doing it amazing and we want to see you grow. So feel free to share your wins and everything down below. With that, we're here ever we're here live every single week, every Friday morning. Uh, we talk everything like this more AI. I'm sure there's gonna be more and more AI and different strategies and impacts that you can do in your business on a regular basis.

Duarne 1:22:05

We'll go a few weeks where we may not we tried not to do that.

Dave 1:22:08

We tried not to, but I think it's just yeah, it's just so impactful. So we're gonna keep talking about it. You know, we've talked finances and stuff like that, which is also important, but I think AI is is generating a lot of as well. So with that, we hope to see you in uh in the next episode next week. If you have any questions ahead of time, feel free to share them down below. And then with have a wonderful week, and we'll see you.

unknown 1:22:33

Bye.