Episode 517: Flip the Script: Start with Your Business Problem, Not the AI Tool

WEBVTT
00:00:00.080 --> 00:00:02.560
So the level of complexity is going to be different for each business.
00:00:02.799 --> 00:00:14.720
But I think that there's a lot of opportunity if you look at it from uh just kind of roadmap, just map out your your you know what that cash, I'm sorry, opportunity to cash flow is within your business today.
00:00:15.599 --> 00:00:19.679
Welcome everyone to another episode of Dynamics Corner.
00:00:20.000 --> 00:00:24.480
Well, AI, where do you even begin as a business leader?
00:00:24.640 --> 00:00:26.239
I'm your co-host, Chris.
00:00:26.480 --> 00:00:27.440
And this is Brad.
00:00:27.519 --> 00:00:31.760
This episode was recorded on April 15th, 2026.
00:00:32.000 --> 00:00:33.520
Chris, Chris, Chris.
00:00:34.000 --> 00:00:35.679
Happy tax day.
00:00:36.240 --> 00:00:38.079
Here in the United States.
00:00:38.399 --> 00:00:40.240
Where do you begin with AI?
00:00:40.399 --> 00:00:41.840
That is a challenging question.
00:00:42.000 --> 00:00:48.159
We see all of the hype about this agent, that agent, this tool, this tool that tool, this framework, this token.
00:00:48.399 --> 00:00:49.920
It's rather confusing.
00:00:50.159 --> 00:00:54.079
But if I have an organization or a business, where do I begin with AI?
00:00:54.240 --> 00:00:56.799
Today we had the opportunity to have that conversation with Max Right.
00:01:06.879 --> 00:01:08.719
Greaters, how are you guys doing?
00:01:09.920 --> 00:01:10.879
Well, how about yourself?
00:01:11.120 --> 00:01:12.239
How about yourself?
00:01:12.480 --> 00:01:12.799
Good.
00:01:12.959 --> 00:01:15.439
I didn't know that uh you were a soccer fan.
00:01:16.640 --> 00:01:17.680
Oh yeah.
00:01:18.480 --> 00:01:19.599
Yeah, big time.
00:01:19.760 --> 00:01:20.400
Yeah.
00:01:20.799 --> 00:01:23.040
Do you do do you follow MLS?
00:01:23.280 --> 00:01:28.000
Or just uh European or or uh National League?
00:01:28.319 --> 00:01:33.599
Uh I I follow both Premier League and um and uh MLS.
00:01:33.680 --> 00:01:36.000
I'm a big FC Cincinnati fan.
00:01:36.640 --> 00:01:37.280
Nice.
00:01:38.079 --> 00:01:42.159
Uh we've had season tickets since they started all the way back in 2016.
00:01:42.319 --> 00:01:43.920
So interesting.
00:01:44.319 --> 00:01:45.840
I'm a New England Revolution fan.
00:01:46.480 --> 00:01:46.719
Too bad.
00:01:47.120 --> 00:01:49.359
Had I known, I would have grabbed my scarf.
00:01:51.040 --> 00:01:51.760
That's great.
00:01:52.400 --> 00:01:54.400
I have season tickets for the revolution.
00:01:54.719 --> 00:02:01.519
I had them when they first, well, when they first started or when they first went to Gillette Stadium that I stopped for several years and I picked them up again.
00:02:01.760 --> 00:02:02.239
Nice.
00:02:02.879 --> 00:02:03.200
Yeah.
00:02:03.359 --> 00:02:07.760
I've held on to mine the whole time, even through three wooden spoon seasons.
00:02:08.080 --> 00:02:08.479
That's good.
00:02:08.639 --> 00:02:09.439
Soccer's fun.
00:02:09.520 --> 00:02:16.560
It's one, it's one of those sports that once you know the rules, oh yeah, it's fun to watch.
00:02:16.719 --> 00:02:22.879
But I think if you were to just pick it up and sit and watch it, it's just a bunch of people kicking a ball.
00:02:23.039 --> 00:02:23.280
Yeah.
00:02:23.599 --> 00:02:24.639
But yeah.
00:02:25.360 --> 00:02:34.639
Once you learn the rules, I got into the rules, and then you get a greater appreciation for the game and the footwork.
00:02:34.800 --> 00:02:34.879
Oh.
00:02:35.199 --> 00:02:40.479
And watching what some of these athletes can do with that ball is amazing.
00:02:40.639 --> 00:02:41.919
Like where they can place it.
00:02:42.000 --> 00:02:43.680
Like, how can you play the setup?
00:02:43.759 --> 00:02:45.199
The setup as easily further.
00:02:45.360 --> 00:02:50.080
Like you watch some of these even like corner kicks or some of these other kids, the ball just literally goes and arcs.
00:02:51.039 --> 00:02:53.360
And I'm like, how can you do that?
00:02:53.520 --> 00:02:58.319
Yeah, at full sometimes at a full speed run, you know, it's it's it's pretty fun to watch.
00:02:58.400 --> 00:02:59.039
But I agree with you.
00:02:59.120 --> 00:03:02.560
I I started, I didn't pay any attention to soccer at all when I was growing up.
00:03:02.639 --> 00:03:03.840
I didn't have much interest in it.
00:03:03.919 --> 00:03:05.759
But my son started playing when he's five.
00:03:05.919 --> 00:03:07.280
He's 22 now.
00:03:07.919 --> 00:03:13.039
And uh I was asked to be an assistant coach a couple of years into that when he was really young.
00:03:13.280 --> 00:03:15.439
And uh, I'm just that kind of guy, right?
00:03:15.520 --> 00:03:18.400
So I when I was asked to do that, I was all in.
00:03:18.560 --> 00:03:23.439
So friends of mine are trained coaches, they gave me recommended reading.
00:03:23.759 --> 00:03:35.039
I went headlong into it, really understood the game, and uh coach coached him for about three years as an assistant coach, and those were his only three winning seasons.
00:03:35.120 --> 00:03:36.879
I'm not gonna take all the credit.
00:03:38.960 --> 00:03:42.639
It was a lot of fun, and that's really when I fell in love with the game, and then we got it.
00:03:43.039 --> 00:03:55.439
It's it is a great sport, yeah, and to watch some of these athletes, like you said, what they can do with the ball at full speed and with the footwork, it's almost like the ball is part of their body.
00:03:55.520 --> 00:04:05.439
It's a hard thing to explain or understand, but it's almost as if they maneuver and kick and and handle that ball as if it's one of their hands.
00:04:05.680 --> 00:04:09.280
They just move it like uh I just yeah, listen.
00:04:09.680 --> 00:04:18.639
A friend of mine, he's uh so my son, when he got to about 10 years old, he started playing select, and then I wasn't, you know, I'm not I couldn't coach at that level.
00:04:18.959 --> 00:04:22.639
He needed he needed much better expertise and opportunity.
00:04:22.879 --> 00:04:33.759
And uh he joined a club here in Cincinnati that was just getting off the ground, and one of the two owners is a British gentleman who played at a pretty high level, not Premier League, but he played at a high level in Europe.
00:04:34.319 --> 00:04:41.279
And um, we're still friends, and he just posted a highlight reel of his son that's probably seven years old at this point.
00:04:42.240 --> 00:04:46.160
This kid the highlight reel was insane.
00:04:46.399 --> 00:04:48.160
He's dribbling right at everybody.
00:04:48.399 --> 00:05:04.079
He one of the highlights, he brings the he brings the ball up behind him, you know, kind of brings the ball back when the defenders collapse in on him, brings it over his back shoulder, pops around these two guys, and then one touches it into the goal right above the the keeper, top of the net.
00:05:04.240 --> 00:05:06.240
I'm like, he's like eight, maybe.
00:05:06.399 --> 00:05:07.759
I'm like, what are you doing?
00:05:09.839 --> 00:05:16.079
It's it's uh it's a great sport, it's interesting, but uh uh I I could talk soccer all day.
00:05:16.240 --> 00:05:19.600
Uh unfortunately, uh time doesn't permit that.
00:05:19.680 --> 00:05:22.399
But before we get into the conversation, might tell us a little bit about yourself.
00:05:22.639 --> 00:05:23.199
Yeah, sure.
00:05:23.279 --> 00:05:23.439
Yeah.
00:05:23.519 --> 00:05:26.560
So Matt Stribble Hoff, CEO of Red Hawk Technologies.
00:05:26.800 --> 00:05:28.879
Grateful to be back on the podcast, guys.
00:05:28.959 --> 00:05:30.480
I always love our conversations.
00:05:30.639 --> 00:05:34.720
It's the energy is always up there, and we talk about some really cool, crazy stuff.
00:05:34.879 --> 00:05:38.160
Um, but we just celebrated our 18th birthday in business.
00:05:38.319 --> 00:05:41.519
And uh wow, happy birthday! Yeah, thank you, man.
00:05:41.759 --> 00:05:42.480
Really proud of that.
00:05:42.639 --> 00:05:48.079
It's it's uh it's it's great to get past that first couple of years, that first five-year milestone.
00:05:48.160 --> 00:05:53.199
A lot of businesses don't make it there, and now here we are at 18 and growing strong.
00:05:53.519 --> 00:06:04.639
And uh, you know, I went from being really concerned and worried about AI and its impact on software engineering to wholly embracing it and kind of geeking out.
00:06:04.720 --> 00:06:07.439
So I'm excited to kind of share where we are now.
00:06:07.920 --> 00:06:09.519
It is easy to geek out.
00:06:09.759 --> 00:06:23.439
I think that in my opinion, uh, there's been a lot of excitement about it, but I'm uh there still continues to be a lot of excitement, a lot of features, functionality added to the AI models, yeah.
00:06:23.519 --> 00:06:30.399
Uh from the software development point of view, uh, what you can do with it, but also from the business point of view.
00:06:30.560 --> 00:06:35.680
Yeah, a lot of people think of AI, artificial intelligence, we always talk about it's a broad topic.
00:06:35.839 --> 00:06:37.519
You know, what do you mean by AI?
00:06:37.680 --> 00:06:45.040
It's not just coding, it's not just uh um, you know, chat GPT and asking questions.
00:06:45.199 --> 00:06:49.120
It's uh it's it's a you can use it as an assistant.
00:06:49.199 --> 00:06:53.360
Uh I use it as an assistant now, more so than anything.
00:06:53.600 --> 00:07:00.399
I do a lot of coding with it, but I use it also as more of a personal assistant to do research, to uh tabulate things.
00:07:00.480 --> 00:07:05.680
I've been looking at some properties and I go around and I have AI just put together a nice portfolio of the properties for me.
00:07:05.839 --> 00:07:16.560
I talk to him, I look at them, I speak my thoughts, it asks me questions and makes nice sheets about each property and then compares them to the others and it'll do research on it.
00:07:16.720 --> 00:07:24.720
And to be honest with you, it asks me questions about them as I'm viewing them that I wouldn't even think of looking at in some cases.
00:07:24.879 --> 00:07:53.360
So yeah, my business partner right now is negotiating with uh, I guess it's it's the the the folks putting in the new um gas pipeline, and he's gonna lose, he has to give up certain percentage of his property, and he's used it to help him through that whole negotiation process, and and then he puts it, cycles things back through, and they're like, Oh, they're gonna know that you have a certain level of sophistication if you approach it this way, and so you might want to kind of start here, and it's just it's nuts.
00:07:53.519 --> 00:07:58.639
Wow, it actually asked him questions around uh old timber.
00:07:58.720 --> 00:08:03.519
Like, do you are you gonna or are there any trees that have to come down along this easement?
00:08:03.759 --> 00:08:05.759
And he's like, Well, yes, and okay, well, what are those trees?
00:08:05.839 --> 00:08:07.199
And he says he's communicating with it.
00:08:07.279 --> 00:08:14.160
It's like, okay, you've got some old growth hardwood there that's probably those trees might be worth, you know, twenty-thirty thousand dollars.
00:08:14.240 --> 00:08:16.079
You gotta add that into your plan when you're negotiating.
00:08:16.240 --> 00:08:19.439
It's like it's it's scary.
00:08:19.519 --> 00:08:25.439
I mean, it again, it's sometimes uh from my experience, I've seen it do things really, really well.
00:08:25.600 --> 00:08:26.079
Yeah.
00:08:26.399 --> 00:08:30.720
And I've seen it do things really, really bad.
00:08:30.879 --> 00:08:31.199
Yeah.
00:08:31.439 --> 00:08:39.440
I I think you st it's still something that you need to review uh uh uh at certain points.
00:08:39.519 --> 00:08:41.759
And it's hard, uh I think you'll understand what I'm saying.
00:08:42.080 --> 00:08:50.799
You get a sense of where you need to have that human in the loop type process, uh, the more you work with it and and how to work with it.
00:08:51.120 --> 00:08:53.279
And uh we do appreciate you coming back.
00:08:53.600 --> 00:08:54.639
Always nice to have a follow-up.
00:08:54.799 --> 00:08:56.240
You're you're big in the AI world.
00:08:56.480 --> 00:09:16.080
I was thinking uh uh of catching up on some of the things that you're doing with AI, but I I've had a couple of conversations this week, and I I thought that you would be someone that would be able to help navigate some of the questions that I'm hearing from other uh individuals with businesses.
00:09:16.399 --> 00:09:16.960
Sure.
00:09:17.600 --> 00:09:25.360
And the first one is again, everybody's at a different stage of their journey with AI.
00:09:25.519 --> 00:09:34.559
Uh everybody hears about it because I don't think you can walk one step on this planet without AI being thrown at you in some form.
00:09:34.960 --> 00:09:36.000
Uh the word AI.
00:09:36.080 --> 00:09:40.399
Now I'm not saying like use of AI, I'm saying the word AI being thrown at your face.
00:09:41.200 --> 00:09:53.519
But I've had conversations with the business team members and and uh uh business leaders uh over the past week and last week as well.
00:09:55.360 --> 00:09:57.039
Where do you begin?
00:09:57.440 --> 00:09:57.759
Right?
00:09:57.919 --> 00:10:00.879
This is one of the I think a lot of people are getting into it.
00:10:00.960 --> 00:10:02.240
Okay, well, this isn't a hype.
00:10:02.320 --> 00:10:08.000
At first, I think you have people like, okay, this is just a hype, or oh, this is just chat GPT, I can ask you questions, it gives me answers.
00:10:08.240 --> 00:10:14.240
And then other then you start seeing more and more of some of the social media hype of what people do with AI.
00:10:14.320 --> 00:10:25.200
Oh, I have 4,000 agents working for me, running my business on seven Mac minis, and I'm making$4.2 billion a day, all the way down to you know, but I was saying as you see how there's a type of exposure, sure.
00:10:25.360 --> 00:10:34.480
So I've had these conversations in these groups, and that was a very good question of what is AI in my business.
00:10:34.639 --> 00:10:35.840
How where do I begin?
00:10:36.000 --> 00:10:39.919
How do I even know what I want to do with it?
00:10:40.639 --> 00:10:42.399
And how do I assess it?
00:10:42.559 --> 00:10:45.200
And how do I get into it, and how do I get started?
00:10:45.360 --> 00:10:57.759
And I know you have your business uh developing software or creating software applications, and you was our last episode we talked about, you have a big use of AI within your organization.
00:10:57.840 --> 00:11:01.840
We can talk about that as well, but I'm thinking, how does someone get started in this journey?
00:11:02.080 --> 00:11:11.039
Well I think that it's very much along the lines of traditional business strategic planning.
00:11:11.519 --> 00:11:23.519
If, for example, take AI completely out of the conversation and then look at your business holistically, you're gonna have to have your business leads in this strategic planning conversation.
00:11:24.159 --> 00:11:39.600
But where are your most expensive workflows within that business where you can streamline operations and help drive more revenue to the bottom line just by simply trying to reduce some of the cost of the way that you're operating?
00:11:40.399 --> 00:11:46.080
There's a lot of ways to reduce those types of workflows, you know, the operational cost of those workflows.
00:11:46.559 --> 00:11:55.759
It may end up being machine learning and you know, data warehousing and things like that to streamline operations, or AI might fit into the conversation agentically.
00:11:55.840 --> 00:12:09.200
Maybe there's some things that can be done there to help assist uh the people that are critical and the subject matter experts particularly that are critical in making sure that that part of your business operates efficiently.
00:12:09.600 --> 00:12:14.159
But if you if you just take AI out of the conversation, it might remove some roadblocks.
00:12:14.320 --> 00:12:21.360
People think they have to start with how you know AI, where do I plug it in in order to receive some benefit?
00:12:21.679 --> 00:12:23.200
I would say flip the script.
00:12:23.440 --> 00:12:31.440
Look at your business holistically first and think about uh where are your most expensive workflows within the organization?
00:12:31.519 --> 00:12:41.279
And expense might be a result of, if you're in a highly regulated industry, how often are we failing or struggling with specific types of third-party audits?
00:12:41.360 --> 00:12:43.039
What does that cost our organization?
00:12:43.120 --> 00:12:47.679
What is causing those uh mistakes that are leading to that expense?
00:12:47.840 --> 00:12:54.879
And can an agentic workflow or tooling help us reduce the amount of occurrence or frequency of occurrence of those mistakes?
00:12:55.120 --> 00:13:00.879
But really what you're doing is you're talking about the business challenge first to find the opportunities.
00:13:01.039 --> 00:13:04.240
And so we actually created an AI decision matrix.
00:13:04.320 --> 00:13:05.600
It's really pretty simple.
00:13:06.000 --> 00:13:20.960
And you know, people can download it on our website if they're interested, but it's a way to kind of reframe the conversation so that you're not so focused on AI's capabilities and all the hype and the buzz out there and and how you replicate that type of success in your business.
00:13:21.200 --> 00:13:28.080
Just I would suggest people take that off the table and actually start with what your core business functions are.
00:13:29.519 --> 00:13:35.840
Where is there a lot of um transactions with regard to data and people?
00:13:36.000 --> 00:13:40.000
Because you know, you if you're talking about opportunity to cash, right?
00:13:40.080 --> 00:13:42.320
Every business is opportunity to cash.
00:13:43.360 --> 00:13:44.000
Really?
00:13:44.320 --> 00:13:45.120
No, I'm just kidding.
00:13:45.279 --> 00:13:45.759
Yeah, right?
00:13:45.919 --> 00:13:50.559
I mean, fundamentally, it's I think people forget that people forget that point.
00:13:50.799 --> 00:13:53.279
All businesses are in business to make money.
00:13:53.440 --> 00:13:54.240
Yes, period.
00:13:54.720 --> 00:13:58.480
It's it's there is no other reason for it.
00:13:58.559 --> 00:14:08.159
I mean, I it I'm not going to get down the whole nonprofit role or the whole nonprofit discussion, but all businesses are in it to make money somewhere for someone.
00:14:08.399 --> 00:14:18.399
And it it you're going to have business profitability, which business profitability is going to go to its employees, and you're making money for those employees so that they can live.
00:14:18.480 --> 00:14:23.200
So it's the fundamental existence of a business is not just to have fun.
00:14:23.440 --> 00:14:29.600
Uh, people may have fun while they're doing it, but the underlying goal is to be profitable.
00:14:29.840 --> 00:14:30.320
Absolutely.
00:14:30.480 --> 00:14:33.919
And and part of that is reducing the time to cash.
00:14:34.639 --> 00:14:34.879
Right?
00:14:35.039 --> 00:14:40.000
How do you compress the time from opportunity to actually having that revenue hitting the books?
00:14:40.559 --> 00:14:48.879
And in every business, you're going to have different leaders that are responsible for different functions within that the way that you're you're structured operationally.
00:14:49.039 --> 00:15:02.240
So there are all these handoffs that take place from marketing, you know, qualified lead to the sales team, to now there's an intake process for that new deal that you just landed.
00:15:02.799 --> 00:15:07.759
And every company has different ways that they slice that up, particularly you know, based on what they're offering.
00:15:08.000 --> 00:15:09.600
You know, is it an engineered product?
00:15:10.000 --> 00:15:10.799
What is it, right?
00:15:11.039 --> 00:15:13.679
So the level of complexity is going to be different for each business.
00:15:14.080 --> 00:15:25.440
But I think that there's a lot of opportunity if you look at it from uh just kind of roadmap, just map out your you know, what that cash, I'm sorry, opportunity to cash flow is within your business today.
00:15:25.919 --> 00:15:31.679
Who are the subject matter experts responsible for those successful handoffs from department to department?
00:15:33.279 --> 00:15:36.720
Where where are the areas where you know that you have some inefficiency?
00:15:37.360 --> 00:15:43.919
Now look at those inefficiencies and see how AI or ingenic workflow or some type of integration might help optimize that.
00:15:44.639 --> 00:15:46.320
I don't think that it's rocket science.
00:15:46.480 --> 00:15:47.200
I really don't.
00:15:47.679 --> 00:15:55.600
I I I like that approach because I think breaking it down to look at specific tasks and seeing how AI may be able to assist you.
00:15:55.759 --> 00:15:59.039
And you reference uh a couple scenarios there.
00:15:59.200 --> 00:16:04.720
You reference what's your most expensive, but where are you where do you have an expensive process?
00:16:04.960 --> 00:16:05.519
Yes.
00:16:05.840 --> 00:16:10.240
And now you're talking about take a look at that operations to cash.
00:16:11.120 --> 00:16:22.639
And in that, I'm assuming or inferring from your conversation that expense doesn't necessarily always mean dollars when you're looking at the expense, it could be time.
00:16:22.879 --> 00:16:23.120
Yes.
00:16:23.279 --> 00:16:31.840
So in some of these operations, what is the friction you may have that will delay your opportunity to cash cycle?
00:16:32.080 --> 00:16:39.519
And how may AI help you reduce that friction, either cutting or reducing steps?
00:16:39.759 --> 00:16:50.559
Uh, because uh I I know, and I'm happy that you brought that up because again, these are discussions that I've had with uh every time I'm up north, I go to dinner with many different people.
00:16:50.720 --> 00:16:55.679
It's strange, and I I think I gain like 300 pounds when I'm here, but that's a different story.
00:16:55.840 --> 00:17:00.080
Um, because all we do is go out to eat and um drink water.
00:17:00.480 --> 00:17:11.759
So but it is, it's it's the it's the friction, and anytime you hand off to a person, what is the delay in the feedback loop as well?
00:17:12.000 --> 00:17:21.920
Because it could be, as you had mentioned, you hand it from a sales team to uh an implementation team or a sales team to an engineering team for architecture.
00:17:22.160 --> 00:17:28.880
There's a feedback loop between those in order to continue to the next step or to go to the next step.
00:17:29.359 --> 00:17:46.720
So maybe looking at some of those handoffs to see how can you make that more efficient from a time point of view, which in essence increases your opportunity to cash as well as reducing cost of time.
00:17:47.759 --> 00:17:59.680
And if you're in consulting, sometimes that time can be expensive because it could be the opportunity of uh being successful or not successful in winning an engagement as well.
00:18:00.319 --> 00:18:00.559
Certainly.
00:18:00.799 --> 00:18:02.880
And I and I think there are there are a lot of great tools.
00:18:02.960 --> 00:18:05.440
You guys are clearly you know familiar with notebook.
00:18:05.920 --> 00:18:07.680
You know, Copilot has their version of that now.
00:18:07.759 --> 00:18:11.279
I think in a prior episode I was talking about you know Google's version of that.
00:18:12.079 --> 00:18:28.480
If you can capture all your transcripts and meetings, conversations, content in a notebook for an engagement from lead to cash and interview each of the parties responsible to evaluate where your what your current status is.
00:18:28.559 --> 00:18:34.559
Because your subject matter expert, if your boots on the ground, those are the folks who know where all the problems and challenges are.
00:18:35.759 --> 00:18:47.839
And so one thought is to implement a strategic you know, information gathering, evaluation process, leveraging AI, and using notebook as as really your research assistant.
00:18:48.799 --> 00:19:13.680
So if you're not quite sure where some of those challenges or opportunities lie within that flow from lead to cash, follow some of that through your the process, capture all of your interviews with the individuals responsible, dump all that into a notebook and then start vibing with that AI agent and see what kind of ideas come from leveraging that as a research assistant.
00:19:15.359 --> 00:19:20.480
Okay, so that's that's a good approach to start, and that's uh where do we start?
00:19:20.640 --> 00:19:21.039
Yes.
00:19:21.440 --> 00:19:25.440
And now how do we start?
00:19:25.680 --> 00:19:25.920
Right?
00:19:26.079 --> 00:19:30.000
So now it's we've we've done, we've identified the where, which could be the how, right?
00:19:30.079 --> 00:19:34.640
But the how is to start is to do the the review of these processes.
00:19:34.880 --> 00:19:44.319
But now I hear you had mentioned notebook Google has uh perplexity, notebook LM, uh, which has some nice features in it.
00:19:44.480 --> 00:19:56.000
We hear uh on the news open AI, chat beat, chat GPT, codex, claude, uh, copilot, opus, sonnet, haiku, I can keep going.
00:19:56.079 --> 00:20:03.279
It's it's it's there's there's a a smorgasborg of words that we hear with AI.
00:20:04.000 --> 00:20:12.079
And if I'm new to it, or I'm really trying to understand it, because there are businesses that aren't technical businesses.
00:20:12.160 --> 00:20:15.359
We all work in a technical space, so we're in this bubble.
00:20:15.440 --> 00:20:18.960
I'm starting to realize of understanding and using and knowing AI.
00:20:19.039 --> 00:20:41.839
Yes, but I speak with individuals outside of the tech sphere that have businesses doing HVAC that doing that service other service type businesses or uh um commercial type businesses, and they're not so exposed or as exposed, and again, it's it's okay because everybody has to start at a different point in their journey.
00:20:42.000 --> 00:20:53.200
They're not so exposed to the technical workings of it, and some of them are rather afraid because they hear about all this stuff and they hear all these words, and they're like, Well, what should I know?
00:20:53.359 --> 00:20:58.559
Like, how do I even know what AI can do, and how do I even set it up and how do I go through this?
00:20:58.720 --> 00:21:01.519
So, what's your take on that?
00:21:01.759 --> 00:21:13.200
I know it's it's kind of putting you on the spot a little bit, but uh Chris, jump in as well, too, from your thoughts on any of this stuff, because this is these are real conversations I've had more than once in the past week.
00:21:14.000 --> 00:21:27.519
Well, I I think I think sometimes too that uh the fear of look, when when AI was coming around and everyone's like, oh, it's gonna take a job, even that alone prevents people from starting.
00:21:27.759 --> 00:21:35.920
Like, why would I want to make an effort if it's if there's a possibility that's going to take my job, I'm not gonna go and do that.
00:21:36.160 --> 00:21:46.160
So I think that's a I think that's a valid challenge right now where where maybe the adoption is a little slower in other areas or other industry.
00:21:46.559 --> 00:21:52.000
Um I think for us it's easy to say, oh, we should adopt it because we are in tech.
00:21:52.319 --> 00:21:55.119
But there are other industries where, like, I don't even know where to start.
00:21:55.200 --> 00:21:55.599
You're right.
00:21:55.680 --> 00:21:57.119
So it's very difficult to understand.
00:21:57.279 --> 00:21:59.759
You can have some ideas, but then you get to a point.
00:22:00.079 --> 00:22:01.519
Okay, okay, now we have a problem.
00:22:01.680 --> 00:22:02.480
We have a challenge.
00:22:02.559 --> 00:22:04.640
Let's go use AI to solve it.
00:22:04.960 --> 00:22:07.440
But to your question, Brad, where do you start?
00:22:07.599 --> 00:22:08.319
Where do you begin?
00:22:08.400 --> 00:22:09.839
Which one do you choose?
00:22:10.079 --> 00:22:12.000
And that is where you get hung up as well.
00:22:12.160 --> 00:22:13.920
And say, okay, well, I don't know which one to choose.
00:22:14.160 --> 00:22:16.079
Someone tells me that this is better for that.
00:22:16.240 --> 00:22:19.279
Another person tells me this one's better for this.
00:22:20.000 --> 00:22:27.359
And making that choice is uh as a business leader, it's like, well, um, where's gonna be my ROI?
00:22:27.519 --> 00:22:31.519
If it's not the right one, uh, did I just waste money?
00:22:31.680 --> 00:22:34.559
Uh so analysis paralysis.
00:22:35.200 --> 00:22:37.359
Absolutely, you know.
00:22:37.599 --> 00:22:42.559
So I I I believe firmly at this point, you know, it is it is absolutely an arms race, you know.
00:22:42.799 --> 00:22:52.640
And and I think it's less about which tool you choose, and it's more important just to choose to start, knowing that it's okay to change direction.
00:22:54.160 --> 00:23:02.640
But when you decide to start, and you choose, let's say that you're gonna start with Chat GPT uh or perplexity.
00:23:03.440 --> 00:23:05.519
There's nothing wrong with starting with either one of those.
00:23:05.680 --> 00:23:14.960
I think there are some nuances, one that you know, things that we as experts might think, well, I'll I'll prefer to use this because of the model behind it, or et cetera, for certain types of outcomes.
00:23:15.359 --> 00:23:26.480
But when we're talking to non-technical business leaders in HVAC, for example, that are are interested in just trying to understand where to start, I would say choose whichever one that you feel most comfortable with.
00:23:26.559 --> 00:23:29.519
And if you're not sure which one that is, just choose one.
00:23:31.119 --> 00:23:35.519
But do so in a way that you're creating an account specific to your business.
00:23:36.000 --> 00:23:37.200
It's not a personal account.
00:23:38.079 --> 00:23:46.319
Make sure that the information that you share with it about your business and processes, look take the time to look at the privacy policy in terms of use.
00:23:46.400 --> 00:23:48.880
Make sure it's not training on the information you feed it.
00:23:49.359 --> 00:23:56.880
I think as long as you can validate that it's uh a tool you can use that's not going to train on your data, then it's okay to choose that tool.
00:23:58.559 --> 00:24:00.319
Yeah, I that's a good point.
00:24:00.400 --> 00:24:03.119
Uh you had mentioned about policy as well.
00:24:03.359 --> 00:24:08.720
You know, before you even begin, you know, what are what are your company policy about AI use?
00:24:09.279 --> 00:24:16.559
Because you have some IP, you have some process that's unique to you, um, that could be it could be exposed.
00:24:16.720 --> 00:24:24.079
I think finding a good partner that can talk you through that is very important.
00:24:24.559 --> 00:24:38.400
I know there's a lot of a lot of businesses want to just turn it on uh and not have any clue of like, okay, what there's a risk there uh in terms of security of your data.
00:24:38.640 --> 00:24:39.839
Uh I've heard that too.
00:24:39.920 --> 00:24:44.559
I've heard stories about people turning it on because they just want to turn it on.
00:24:44.720 --> 00:24:54.960
Someone um uh uh someone had bought or purchased Chat GPT licenses uh for individuals, not the enterprise.
00:24:55.119 --> 00:24:56.640
And it's like use it.
00:24:57.119 --> 00:24:59.519
It's like okay, that's a wrong approach.
00:25:00.000 --> 00:25:04.400
Well, that's that that's another uh challenge in itself.
00:25:04.480 --> 00:25:11.200
And I think we hit a couple of points of where do we start and how do we choose uh Chris?
00:25:11.279 --> 00:25:18.640
You had mentioned having an AI policy, and again, uh these are all things from you think of someone outside of this, how does it start?
00:25:18.799 --> 00:25:34.480
And you're also a point of there's a big difference between using Chat GPT, Claude, Copilot, uh even Grok at home saying, Oh, tell me about my shopping list or uh what happened with the sports game last night.
00:25:34.640 --> 00:25:52.799
You know, I mean there's a big difference between using it for personal use, then bringing it into the business use, and maybe not having I don't like to use the word control, but having visibility or having that human in the loop type process to make sure that the use of it's in the best interest for your business.
00:25:53.119 --> 00:25:57.519
I think shadow IT is becoming more of a problem.
00:25:57.759 --> 00:26:32.160
Not to again, I'm trying not to jump all over the place, but uh uh Chris, you're talking about somebody taking a personal chat GPT subscription into the workplace to start doing things for work, where there may not be a policy or there may not be guardrails, and then they start creating these shadow IT processes or or shadow IT workflow systems, and now an organization can be cut shaught off guard by something that's being done by somebody who's trying their best just to do their job.
00:26:32.319 --> 00:26:42.400
Not to say that someone's doing this stuff maliciously, uh oftentimes uh their intent isn't malicious, even though it could the results can be catastrophic.
00:26:42.960 --> 00:26:53.119
So yes, policies are critical, and then having an owner of that policy and then training people is key.
00:26:53.440 --> 00:27:22.319
So uh it's just it's a pretty simple structure, but in our business we have a support group that they own defining policies for the tools that we're adopting, and then distributing that among the organization, and people are tested on their knowledge of that policy before they're onboarded to use those tools to drive enforcement of the policy itself.
00:27:22.880 --> 00:27:26.799
It's really not that complicated, but it is a process that you you have to implement.
00:27:26.960 --> 00:27:33.119
And we're gonna by the likely within the next year we'll have a hundred engineers on our team.
00:27:33.279 --> 00:27:46.720
So we're we're scaling at a point now where if we don't have that type of system and process in place, shadow IT, especially with engineers who are like, oh man, check out this new tool.
00:27:46.799 --> 00:27:49.200
I'm gonna see what I can get out of base 44.
00:27:49.680 --> 00:27:52.400
Oh, I've got some client you know information over here.
00:27:52.480 --> 00:27:56.960
I'm gonna see how quickly I can videcode a prototype for them as part of this next phase of work.
00:27:58.240 --> 00:28:14.079
Well, if we can't have them just randomly onboarding it, and not only is there a risk from uh IP and and uh information kind of getting out there into these models, there's also a cost implication.
00:28:15.920 --> 00:28:21.359
Because of the tokenization and at scale, that starts to get that can start to get expensive.
00:28:21.519 --> 00:28:43.759
But I don't wanna I don't want to get too deep in the technical side of it because I mean, yeah, it's it's it's that's just see this is where I think all of these conversations that I'm having outside of the tech sphere, as I'll call it, yeah, is is concerning businesses because you start to hear of all these small little stories.
00:28:43.920 --> 00:28:59.680
You hear the great stories of oh, I can run my business uh more efficient by myself, and uh everyone else will be out of business to well, my business blew up because I spent$40,000 yesterday on tokens because I didn't put a limit because somebody was trying to do something.
00:28:59.839 --> 00:29:01.759
So it's tough to go down that road.
00:29:01.839 --> 00:29:13.279
But I think just where do you start and just being aware of some of these things to say that it's it's it's easy to get started with a personal subscription.
00:29:13.680 --> 00:29:24.400
Implementing it into a business requires a little more thought than just turning it on, is what I think the both of you were saying, and what I'm seeing as well.
00:29:24.640 --> 00:29:36.160
It's it's okay to dabble with some small things, but before you implement something organizationally, you should identify your focus of what you're trying to do.
00:29:36.319 --> 00:29:40.319
So which what Chris and I like to talk about this a lot when we have our conversation?
00:29:40.480 --> 00:29:42.000
What problem are you trying to solve?
00:29:42.319 --> 00:29:48.559
That should be the first thing you do before you do anything, yeah, is clearly identify the problem.
00:29:48.799 --> 00:29:49.039
Yeah.
00:29:49.279 --> 00:29:55.839
If you're trying to just create something cool and you don't have a problem to solve, it's all wasted investment at that point.
00:29:56.000 --> 00:29:59.599
It's wasted investment, and what I've seen is I've seen it create more problems.
00:29:59.839 --> 00:30:00.240
Oh, yeah.
00:30:00.480 --> 00:30:02.000
Yeah, people tend to play.
00:30:02.079 --> 00:30:04.559
I I think I think we do have to slow down.
00:30:04.640 --> 00:30:08.000
I know we've spoken to a lot of people that are utilizing AI right now.
00:30:08.079 --> 00:30:26.160
You you know, AI is is supposed to be this um accelerator of many things, and what a lot of business leaders should do, um, or at least from my from my perspective, from my in my opinion, humble opinion, uh, we should slow down.
00:30:26.400 --> 00:30:35.440
Uh, I think both of you guys noted you have to identify the challenges, the you know, the the costly operations that you may be dealing with.
00:30:35.599 --> 00:30:43.759
Um in addition to that, um perhaps the um uh policies, uh data.
00:30:43.920 --> 00:30:45.839
Yeah, you have to have good data.
00:30:45.920 --> 00:30:47.759
Uh you have to have clean data.
00:30:48.000 --> 00:31:02.079
And and is it are these data accessible somewhere, or is it someone putting it on a notepad that could help uh some of the you know important component to have good agents to be working for you?
00:31:02.400 --> 00:31:15.920
So there's a lot of things that you have to do pre preliminary preliminary uh prerequisites before turning on or having agents uh implemented in your in your environment.
00:31:16.160 --> 00:31:23.680
So yeah, I'd say slow down a little bit and cover all those basic basis uh in itself.
00:31:25.119 --> 00:31:36.240
So we talked about identifying this policy or excuse me, identifying the problem, coming up with the responsible use of AI type policy.
00:31:36.319 --> 00:31:38.319
Uh I use that loosely.
00:31:38.559 --> 00:31:41.039
I don't know how formal it may or may not need to be.
00:31:41.200 --> 00:31:48.720
Uh, I think each organization can determine that based upon uh their their uh structure.
00:31:50.000 --> 00:31:51.279
What if I need help?
00:31:51.519 --> 00:31:52.720
Like, so this is great.
00:31:52.880 --> 00:31:53.680
This is all great.
00:31:53.839 --> 00:31:55.039
I I I can be a business.
00:31:55.119 --> 00:32:01.279
So we talked about identifying all this, choose a model, start working with it, get a subscription, and look at it.
00:32:02.400 --> 00:32:04.400
How do I get information on what to do?
00:32:04.559 --> 00:32:11.200
Like, is there there's 40,000 AI experts out there?
00:32:11.680 --> 00:32:22.319
Uh, if I'm looking for help, uh, there's uh a lot of information on the internet, which personally now I don't even know what I believe anymore because I see so many different things that sound good.
00:32:22.720 --> 00:32:24.480
And uh sometimes you have to go out.
00:32:24.799 --> 00:32:31.920
Is that even true, or is that just yeah, you know, bait or something, uh just to get a reaction?
00:32:32.079 --> 00:32:35.519
Um and side note, just a little ramble.
00:32:35.680 --> 00:32:46.079
Then you see all these things that people are posting or articles or this that they're creating, that you read it and you're like, there's four billion words and there's no meat because it was all AI generated, but that's a separate story.
00:32:46.559 --> 00:32:50.720
Um how do you get help?
00:32:52.720 --> 00:32:56.720
Well, where do you which resources are available for you to learn about this stuff?
00:32:57.119 --> 00:33:08.480
Well, you can go straight to AI and start getting help there, which is probably you know what people are doing, you know, Google searches and whatnot, and then using perplexity and doing searches there.
00:33:08.559 --> 00:33:11.279
But but uh it's a great question.
00:33:12.160 --> 00:33:38.880
I I think we've all been in the same line of work long enough to know that this trust issue and challenge, when you're asking for expert assistance with something you just clearly don't understand, it almost always leads people through a process of using their network first and looking for referrals to whom they have trusted in the past and had success with.
00:33:39.440 --> 00:33:42.720
That's why the line of work we're in is almost always referral-based.
00:33:44.400 --> 00:33:45.759
It is, it is, yeah.
00:33:46.000 --> 00:33:52.720
I think a lot of business leaders and non-technical, you know, non-technical business leaders are probably doing just that.
00:33:52.880 --> 00:33:55.119
They're they're they're hanging out in their peer groups.
00:33:56.319 --> 00:33:58.160
Maybe it's vestige, I don't know.
00:33:58.240 --> 00:34:01.200
You know, they're they're talking to other people in their circles.
00:34:01.839 --> 00:34:05.839
Um and uh and they're asking that very question.
00:34:06.000 --> 00:34:07.119
How did you go about it?
00:34:07.279 --> 00:34:08.239
Who did you trust?
00:34:08.400 --> 00:34:10.239
What was that experience like?
00:34:11.679 --> 00:34:23.599
I think that's the that's the pathway that most organizations end up going down because of their they're just they don't understand, and it's big, it's scary, it's intimidating, and they want to find a vendor they can trust, a partner who can go in as a trust.
00:34:24.159 --> 00:34:39.039
Yeah, it's it's it's it is difficult, and I think word of mouth, I think a lot oftentimes, again in in the space that we're in and what we're doing is undervalued because sometimes those connections and relationships can drive a lot of business.
00:34:39.280 --> 00:34:47.760
Uh, you also mentioned uh one thing that we haven't mentioned here, I think scale and size of business matter as well.
00:34:48.239 --> 00:35:02.079
Because having somebody who's responsible for your AI policies and implementation and use doesn't necessarily need to be somebody's full-time position if the business doesn't warrant it.
00:35:02.159 --> 00:35:05.920
You can have someone that does that based again.
00:35:06.000 --> 00:35:08.719
I think everything just depends on the scale and the size of your business.
00:35:08.880 --> 00:35:12.239
I think some organizations definitely need it.
00:35:12.320 --> 00:35:12.639
Yeah.
00:35:13.119 --> 00:35:17.519
And it's not a five-minute task, typically.
00:35:17.679 --> 00:35:21.360
With the way things are changing, the way things are evolving.
00:35:21.679 --> 00:35:28.320
There's a little bit of work in the air, and again, not to say that you have to always use the latest as well.
00:35:28.800 --> 00:35:38.559
But things are changing rapidly and understanding how it could be used, the cost of the use, uh, believe it or not, nothing's free.
00:35:38.719 --> 00:36:00.480
It sounds great and free when I can use my you know, I install one of these chat models on my phone, but there's a big scale difference between, again, me looking for a restaurant and a recipe versus somebody trying to gain efficiencies between their sales handoff cycle and their engineering team to do prototypes for prospects.
00:36:00.639 --> 00:36:01.039
Mm-hmm.
00:36:01.280 --> 00:36:01.920
Mm-hmm.
00:36:03.039 --> 00:36:09.760
Most of our customers, and this is sort of the audience that we we serve, are upper mid-market or mid-market businesses.
00:36:10.159 --> 00:36:13.119
We do have S and Bs in our portfolio as well.
00:36:13.760 --> 00:36:31.039
And in my experience, the majority of them rely on fractional CIOs to assist with tech stack evaluation, platform evaluation before they transition from nothing to a new ERP or an ERP they're unhappy with to a new ERP.
00:36:31.199 --> 00:36:36.960
Oftentimes they'll bring in those fractional experts to assist with that type of planning.
00:36:37.039 --> 00:36:50.079
And I know several of these fractional CIOs are also helping them establish acceptance use policies for AI, helping them make decisions based on the current tech stack, what tooling makes sense to adopt, and they're putting some guardrails on that.
00:36:50.559 --> 00:37:03.280
And those are not full-time hires, they're extremely valuable because of the level of expertise they can bring to a smaller organization as a fractional consultant.
00:37:03.360 --> 00:37:14.320
And we're seeing a lot of our customers uh they've been doing that for quite some time, and now those fractional CIOs are expected to bring expertise around AI into that conversation.
00:37:14.480 --> 00:37:17.920
If they're not, they're probably not gonna be the fractional CIO for very long.
00:37:19.440 --> 00:37:25.119
Yeah, I think I think asking for help is uh uh first we we established that earlier today, right?
00:37:25.199 --> 00:37:29.199
Like you've you got to figure out what what challenges and and what problems you're trying to solve.
00:37:29.280 --> 00:37:29.519
Yeah.
00:37:29.679 --> 00:37:36.000
And uh Matt, you had talked about like maybe within your network, uh, talk to the people, your peers that you've interacted with.
00:37:36.239 --> 00:37:44.320
Um, I do believe that going to conferences would be a good opportunity if you are a business leader that's like, I don't know where to begin.
00:37:44.559 --> 00:37:45.599
I have these challenges.
00:37:45.679 --> 00:37:49.119
So first identify all your challenges and and problems you're trying to solve.
00:37:49.280 --> 00:37:50.320
Talk to your peers.
00:37:50.719 --> 00:37:57.679
They may or may not be in your industry, but they've know they've worked on it uh to solve their problems.
00:37:57.840 --> 00:38:01.599
Uh, I think a good one would be to go to a conference.
00:38:01.679 --> 00:38:06.320
Now you come in prepared, obviously, with your questions and challenges and all that.
00:38:06.639 --> 00:38:17.440
You go to a conference, you could be sitting next to someone that have maybe have solved a similar issue that you're dealing with and say, okay, well, let's pick that person's brain.
00:38:17.599 --> 00:38:26.800
Uh, I think it's more, I think it's very important now to go to these conferences because it's not it's no longer just generative AI where we just get answers and stuff like that.
00:38:27.039 --> 00:38:29.440
How are our businesses using agents?
00:38:29.519 --> 00:38:31.199
And I think conferences would be a good place.
00:38:31.280 --> 00:38:42.480
And then of course you get someone speaking about specific agents, but the people that would be sitting next to you uh at these conferences would would would could could help you for sure.
00:38:42.719 --> 00:38:44.400
Yeah, that's a great suggestion.
00:38:44.480 --> 00:38:49.920
Uh I would add that a lot of the conferences include breakout sessions, so it's not always just the keynote.
00:38:50.079 --> 00:39:05.440
And you might find if you're looking at the agenda for the conference even before you sign up, if there are some breakout sessions where they're doing more workshop type planning and and and experience shares, that's oftentimes where a lot of ideas and and networking, etc., can kind of expand.
00:39:07.039 --> 00:39:07.519
Yep.
00:39:07.760 --> 00:39:08.400
Yeah.
00:39:09.119 --> 00:39:10.400
There's a lot to this.
00:39:10.639 --> 00:39:21.519
Now I'm curious to hear a little bit how your AI um use has uh progressed and grown since we last spoke.
00:39:21.760 --> 00:39:23.679
But is it true?
00:39:23.840 --> 00:39:29.199
Can you run 14 agents 24 hours a day, seven days a week, and make 40 billion dollars?
00:39:30.559 --> 00:39:36.800
Is it if I pulled that out I wouldn't be on the podcast right now, it'd be somewhere in some exotic part of the world.
00:39:37.039 --> 00:39:46.960
Uh I think there's uh I think that I don't think we've seen the world's first billionaire that is running a company with just himself and agents.
00:39:47.119 --> 00:39:48.400
Is it possible?
00:39:48.880 --> 00:39:55.039
I think the billionaires that are out there in the tech industry right now are taking bets as to how long it's going to be before we see it.
00:39:55.679 --> 00:39:57.119
I think a lot of things are possible.
00:39:57.199 --> 00:40:03.039
I have engineers right now that um one I'm thinking about today, we were just talking uh earlier this week.
00:40:03.280 --> 00:40:11.840
He on average has five to seven agents running at any given point in time to expedite the work that he's doing.
00:40:12.239 --> 00:40:12.800
Interesting.
00:40:13.039 --> 00:40:33.440
Now that's it's it's I I talk with people all over, and it's I've learned to dissect what everyone says because I had a conversation with someone the other day, and like, oh, they have 14 agents and they're doing all this, and each agent they have one for documentation, one for testing, one for coding.
00:40:33.599 --> 00:40:42.480
And then they come to find out they just have instructions for single-threaded processes, and they have 14 different sets of instruction files.
00:40:42.639 --> 00:40:47.280
So they're not really doing this 24-sev type process.
00:40:47.360 --> 00:40:50.400
They they they may have those separate agents to do separate tasks.
00:40:50.559 --> 00:40:51.760
It's geez, I'll even admit it.
00:40:51.840 --> 00:41:06.320
I have several agents that are doing different things based upon what I want to do because I have guardrails that I want the output to follow or things that I want it to focus on so it doesn't go the wrong way.
00:41:06.480 --> 00:41:13.360
Uh, and I say that just because it's just some of the things that I've seen it do when you don't give it a few little boundaries, have been uh comical.
00:41:14.159 --> 00:41:22.800
So if someone can it's not the agent that's those individuals that are doing well, I think you can do more with less.
00:41:22.880 --> 00:41:30.079
Yeah, but it also still takes someone to have that creativity and it still takes some people to do something.
00:41:30.239 --> 00:41:37.440
It's not, I'm going to say, let's create an agent and create an application, and I don't have to sit at it now.
00:41:37.519 --> 00:41:45.840
I'm on my beach in Tahiti, you know, sipping a margarita while my my team's back there working and running, and uh it's I laugh.
00:41:45.920 --> 00:41:56.960
And and Chris, you said it's what'd you say it's for uh I don't want to say influences, but it's more for like uh I don't want to say clickbait or more for it's just attention.
00:41:57.119 --> 00:41:59.280
Like some of us really would like to see what they're doing.
00:41:59.360 --> 00:42:01.599
I could set up 14 agents to run and do something.
00:42:01.760 --> 00:42:02.320
Sure, you could.
00:42:02.480 --> 00:42:03.039
What is the output?
00:42:03.280 --> 00:42:04.480
Is it gonna be any money?
00:42:04.719 --> 00:42:05.199
I don't know.
00:42:05.280 --> 00:42:06.800
Is it gonna cost you a lot of money?
00:42:07.039 --> 00:42:07.840
Probably.
00:42:08.239 --> 00:42:11.519
Well, the thing is like I I can spend$20,000.
00:42:11.760 --> 00:42:19.760
I like when I see that stuff, I'm like, okay, you're doing all this stuff, but you're spending okay, you have a Claude Max times 20 subscription.
00:42:20.000 --> 00:42:20.639
Great.
00:42:21.280 --> 00:42:22.480
What is that making you?
00:42:22.639 --> 00:42:24.239
Like you can do a lot of cool stuff.
00:42:24.320 --> 00:42:32.400
I'll be the first one to say you can do a lot of cool things and a lot of great things, but how are you capitalizing on it?
00:42:32.639 --> 00:42:34.559
Uh, is the other question that I ask.
00:42:34.639 --> 00:42:39.440
So what value is it creating for somebody else that they're willing to pay you for it?
00:42:40.400 --> 00:42:48.159
Yeah, there are a lot of influencers that are I have seen that some of those uh short reels that they put out there that were like, oh, you could do all this stuff.
00:42:48.239 --> 00:42:52.559
And like you said, Matt, like base 44, you can make an app.
00:42:53.119 --> 00:43:00.880
I think one of the ads were like you can make an app uh and it's an employee within a company that someone used base 44 to create an app.
00:43:01.119 --> 00:43:09.119
I was like, wow, that is uh shadow IT in itself, uh creating something for their company that he was they just happen to use.
00:43:09.280 --> 00:43:18.639
Uh, but there are a lot of influencers out there, and and and unfortunately, some of that that you see somebody see that from a personal point of view, and it's like, well, I could do that.
00:43:18.800 --> 00:43:19.599
Why can't I do that?
00:43:19.679 --> 00:43:22.639
And then they would take it to work, and that becomes a problem, right?
00:43:22.719 --> 00:43:37.280
So a lot of the the uh private or personal side of things uh kind of leaks into the business world, and as a business leader, you that is something is a uh a concern, that is a risk in itself.
00:43:37.440 --> 00:43:39.280
Like, how do you control that?
00:43:39.599 --> 00:43:41.119
Yeah, it's gonna be tough.
00:43:41.280 --> 00:43:43.280
It is gonna be to expect to do that over.
00:43:43.760 --> 00:43:44.559
It is gonna be tough.
00:43:44.719 --> 00:43:53.119
And it then and oftentimes that type of of clickbait and chatter and buzz creates very unrealistic expectations in the business world.
00:43:53.280 --> 00:43:53.920
Yeah.
00:43:54.480 --> 00:43:56.800
Have you been experiencing that with your customers?
00:43:56.960 --> 00:43:59.760
Yeah, like they can see have you been challenged.
00:44:00.159 --> 00:44:08.239
Or questioned by your customers because they see the AI hype or they see the AI slop, I call it.
00:44:08.400 --> 00:44:13.199
Uh, with not just the output of AI, but also the AI slop content that's created on the internet.
00:44:13.599 --> 00:44:23.199
Where I think some may have, in my opinion, a perception of devaluing good output from others.
00:44:23.360 --> 00:44:26.320
Have you had any challenges with that within your organization?
00:44:26.639 --> 00:44:35.360
Well, I think rightfully so, our clients are challenging us to help convey to them how we are adopting AI for their benefit.
00:44:35.519 --> 00:44:38.480
And I think that's fair for them to be asking questions.
00:44:39.440 --> 00:44:42.000
Where does AI fit within your SDLC?
00:44:42.880 --> 00:44:54.079
Uh we have a lot, most our clients are for the most part very reasonable and uh have a high level of sophistication, and so they're not buying into all of the hype, thank goodness.
00:44:55.840 --> 00:45:05.840
But they do expect to see better quality, uh faster throughput from our team as a result of our careful adoption of AI.
00:45:05.920 --> 00:45:10.800
So it's up to us to educate to them as to how we're using it for their benefit.
00:45:11.119 --> 00:45:12.559
And we are doing a lot of that.
00:45:12.719 --> 00:45:21.760
So you know, a lot of that stuff is coming in into the conversation around uh quality assurance testing, automated QA.
00:45:22.480 --> 00:45:31.599
Um, you know, how are we leveraging AI to help us create and then run those tests to reduce the labor and improve the quality throughput?
00:45:31.679 --> 00:45:33.599
So yeah, we're we are having those conversations.
00:45:33.760 --> 00:45:44.800
Now there are occasions where it's maybe it's not a customer, but it's somebody that has been introduced to me that has a brand new idea that might have very unrealistic expectations because of the hype out there.
00:45:44.880 --> 00:45:48.320
And then I then it's I don't know if this is gonna be a good fit, man.
00:45:48.880 --> 00:45:54.320
I can't build 14 agents and and see you on the beach in a month, you know, making 40 million dollars a year.
00:45:54.639 --> 00:46:00.719
Yeah, but when it comes to anything like that, I always tell everyone if it were that easy, everyone could do it.
00:46:00.960 --> 00:46:01.519
That's right.
00:46:01.760 --> 00:46:05.119
And everyone would be on the beach, yeah, and nobody would be working.
00:46:05.360 --> 00:46:14.559
It's it I say that with uh you know, sort of tongue in cheek in a sense, because there are some people, again, you look at Facebook, you know, can anybody create Facebook?
00:46:14.719 --> 00:46:14.960
Yes.
00:46:15.119 --> 00:46:18.880
But somebody at the I say it's the right place and right time and the proper marketing.
00:46:19.199 --> 00:46:22.079
It's it's sort of how you get used and how you get adopted.
00:46:22.239 --> 00:46:23.920
Sometimes it's not what you're creating.
00:46:24.000 --> 00:46:31.760
You don't always need to have this fancy whiz-bang, shiny object, as I call it, this complicated technology.
00:46:31.920 --> 00:46:40.159
Sometimes it's just solving a problem for people and people using that tool and having little friction to use that tool.
00:46:40.480 --> 00:46:53.199
Yeah, you know, we uh this is this is gonna this is an interesting story, and it'll be very brief, but I remember working at another software engineering firm early in my career, we're going all the way all the way back to 1999.
00:46:53.599 --> 00:46:55.119
We were working for a group.
00:46:55.280 --> 00:47:00.800
Our client was uh a company that generated financial reports for companies on Wall Street.
00:47:03.119 --> 00:47:20.079
The challenge that the CEO provided to us was to automate the ingestion of data, create a QuickBooks plugin that would automatically generate charts and graphs because the most expensive part of their workflow were people creating charts and graphs to drop into QuickBooks.
00:47:20.239 --> 00:47:29.360
I'm not sorry, not QuickBooks, um uh Quark Express, which was at the time a tool that designers would use to create annual reports and all kinds of reports.
00:47:29.440 --> 00:47:31.280
It's a design tool.
00:47:32.480 --> 00:47:38.719
Well, we did such a good job at that that we overshot the mark.
00:47:38.880 --> 00:47:41.920
He was looking for an incremental reduction in labor.
00:47:42.719 --> 00:47:50.480
We were able, and this has nothing to do with AI, but it was it was more machine learning and then writing a capsulated postscript to drop in these charts and graphs.
00:47:50.559 --> 00:47:52.960
Now the designer still had to come in and add some information.
00:47:54.320 --> 00:47:56.480
But we reduced their labor by 94%.
00:47:59.840 --> 00:48:00.639
Pretty significant.
00:48:00.800 --> 00:48:05.599
So when we presented that to the CEO, the meeting did not go as planned.
00:48:06.400 --> 00:48:06.639
Right?
00:48:06.719 --> 00:48:08.880
This is something else that's kind of interesting.
00:48:09.119 --> 00:48:11.039
Is the market really ready for it?
00:48:11.760 --> 00:48:21.360
Their entire business and the way that they sold and provided their services to their customers was predicated on the labor associated with generating those materials.
00:48:21.440 --> 00:48:28.320
So we basically reduced his revenue, had he taken that all the way to the finish line by 95%.
00:48:31.280 --> 00:48:45.840
And so they decided not to move forward because in order to do that, they would have to retool everything from the sales process through the contracts to make it value-based as opposed to labor-based calculations in order to produce to deliver their product.
00:48:47.920 --> 00:48:56.239
So sometimes, you know, there are opportunities, and probably more so today with the you know AI tooling that's available, to have a dramatic impact.
00:48:57.840 --> 00:48:59.599
But is are you ready for it?
00:48:59.679 --> 00:49:00.960
Are your customers ready for it?
00:49:01.039 --> 00:49:05.599
Are you can you can you offer that without having a detrimental impact on your revenue?
00:49:05.760 --> 00:49:06.320
I don't know.
00:49:06.480 --> 00:49:10.079
I mean, this is a personal experience we had years ago, but it's something that kind of stuck with me.
00:49:10.159 --> 00:49:15.199
It's like is your big idea feasible?
00:49:15.360 --> 00:49:17.039
Well, first you gotta check that box.
00:49:17.199 --> 00:49:18.239
Let's say that it is.
00:49:19.039 --> 00:49:21.039
Is your business ready for it?
00:49:22.639 --> 00:49:23.760
It's a good point.
00:49:24.079 --> 00:49:25.440
I I think it's a good point.
00:49:25.599 --> 00:49:37.519
Anything that's uh I think any idea or any problem that you're going to solve, uh you have to make sure that you're in the position to support the the the outcome or the output of what you're going to do to solve that problem.
00:49:37.760 --> 00:49:43.920
Yeah, because you might hit the mark and then all of a sudden it's like, oh crap, I gotta change my entire business.
00:49:45.599 --> 00:49:48.639
Which which is it's important to do so.
00:49:48.880 --> 00:49:57.280
And I don't think I I think that if you're going to advance, you have to be aware of that and learn how to manage that.
00:49:57.360 --> 00:50:05.840
Because as you were saying, I think a lot of businesses are challenging how are you using these tools to provide me better value.
00:50:06.079 --> 00:50:06.719
Yes.
00:50:07.280 --> 00:50:34.079
And avoiding using these tools because of maybe how it impacts your revenue, may have a different approach of how can I maximize my revenue by using these tools instead of worrying about these tools reducing my revenue because organizations and businesses see and want to see the value that they're getting.
00:50:34.239 --> 00:50:41.039
Because ultimately, I look at it this way uh if I'm going to buy something, it has a value to me.
00:50:41.199 --> 00:50:45.840
It's going to be$100 for me for this microphone, for example.
00:50:46.159 --> 00:50:47.440
Is it worth it to me?
00:50:47.599 --> 00:50:52.400
If it's$200, maybe I don't want that microphone because it's not worth that much to me.
00:50:52.639 --> 00:51:05.119
So that's it's it's that whole middle piece of providing value, and what is that worth to, you know, is is the value that you're providing worth the cost to the customer?
00:51:05.440 --> 00:51:07.199
Therefore, it's going to impact your revenue.
00:51:07.360 --> 00:51:09.119
It's a little long-winded speech that I had there.
00:51:09.599 --> 00:51:40.719
I I have a couple of friends, one in particular that is a high-level consultant, and his business is work, it is focused on helping professional service organizations modify their go-to-market and pricing strategy to make it value-based so they can liberate themselves from yeah, liberate themselves from a labor cost you know, pricing model so that they don't anticipating, you know, this these what's coming, and they're trying to get to value-based.
00:51:40.800 --> 00:51:41.760
They have to get to value-based.
00:51:42.320 --> 00:52:00.000
We, we, we just, man, that's a topic on its own because I am passionate about that, value-based, uh, and uh specifically in the professional services, and where AI has a huge impact on the professional services industry.
00:52:00.480 --> 00:52:03.039
And uh, yeah, that in itself is a separate topic.
00:52:03.119 --> 00:52:14.880
I mean, we I'm sure Brad and I could talk about that, because that is a an area now with AI has um an impact towards.
00:52:15.119 --> 00:52:18.000
And what does that mean in the professional services?
00:52:18.079 --> 00:52:32.719
Because you sell by you go by the hours and all this stuff, but it's the value is the end result to the you pop that cork right there because uh it is, it's a balancing act.
00:52:32.960 --> 00:52:36.239
What is something that can be value?
00:52:36.880 --> 00:52:55.679
Again, if everything's time-based, if you can do things with less friction and faster, then if you're going to just go by the amount of time it takes, you run the risk of reducing your revenue because you're that more uh you're more efficient and you have to do more, or you charge more.
00:52:56.000 --> 00:53:02.800
So if you're going time-based, if again, if we charge by the hour, and in professional services, many industries, not just tech industries.
00:53:02.960 --> 00:53:06.400
It can be legal, it's you know, any consulting type industry.
00:53:07.360 --> 00:53:11.760
If it takes me less time to do something, I'm going to make less money.
00:53:11.920 --> 00:53:17.840
It's it's it's if I keep the same billing rate per unit of measure of time.
00:53:18.320 --> 00:53:31.440
So if my billing rate is$100 an hour, and now tasks that used to take me three hours take 10 minutes, I run the risk of losing that revenue, or I'd have to say that it took me three hours.
00:53:31.599 --> 00:53:36.559
And it's it's it's it's a balancing act.
00:53:36.719 --> 00:53:40.400
So then you need more to get that same quantity of time.
00:53:40.800 --> 00:53:45.440
But ultimately, at the end, the equation is what's the value to the customer?
00:53:45.519 --> 00:53:50.000
Because even on the professional services base, again, some things you have to pay, uh, unfortunately.
00:53:50.719 --> 00:53:59.599
If you tell me it takes 12 hours of your time to do this task, whatever that task may be, is it worth it to me?
00:54:00.800 --> 00:54:07.440
So I can go into the argument of tell me it's 12 hours at$100 an hour, or tell me it's$1,200.
00:54:08.320 --> 00:54:10.639
Ultimately, I'm gonna get to the same spot.
00:54:11.360 --> 00:54:15.920
Now, I think some things you do have to look at the value of what it is worth to somebody.
00:54:16.000 --> 00:54:17.599
And again, it's not overcharging.
00:54:17.760 --> 00:54:19.840
I am not by any means saying overcharge.
00:54:19.920 --> 00:54:26.000
We go to the supermarket, we pay X dollars for a product because we think it's worth it.
00:54:26.800 --> 00:54:27.039
Right?
00:54:27.199 --> 00:54:34.719
We're going to differentiate between maybe name brand and store brand because the value of the product to us.
00:54:35.039 --> 00:54:39.360
Yeah, and the onus of generating margin is on the supplier of that product.
00:54:39.519 --> 00:54:45.920
So if they can find ways to streamline, modify, you know, working with their vendors, et cetera, to maximize margin, we don't care, right?
00:54:46.000 --> 00:54:48.559
We're just there to buy the product and we're okay with the price.
00:54:48.880 --> 00:54:49.280
Exactly.
00:54:49.360 --> 00:54:54.800
And that's I think some people overlook that fact that the custom I a customer was to look at that.
00:54:54.960 --> 00:54:58.480
Now, there are some things that's still going to be time and material based, I think.
00:54:58.559 --> 00:55:02.320
I think professional services need to take a look at how they're doing these things, and I've been saying this for a while.
00:55:02.400 --> 00:55:12.559
And I'm not saying time and materials as people doing is not valid in some cases, because there are some cases where you truly need to be time-based, right?
00:55:12.639 --> 00:55:27.840
So, again, if I'm giving you providing you know training on site, or if I'm sitting with you, again, that's the reality is I'm spending three hours with you, and you want three hours of my time to do something, whatever that may be, you're going to pay for the time that I'm going to sit with you.
00:55:28.000 --> 00:55:35.519
If you want me to deliver something to you, if we can agree on exactly what that may be, then maybe it's more what's the value to me.
00:55:35.679 --> 00:55:37.519
I really don't, you know.
00:55:37.679 --> 00:55:39.679
I I have some, I say this in every episode.
00:55:39.760 --> 00:55:44.800
If I have someone come do work for me at the house, I really do not care how long it takes.
00:55:45.119 --> 00:55:47.199
I care about how much it costs and when it will be done.
00:55:47.360 --> 00:55:48.880
I mean, I do care about how long it will take.
00:55:48.960 --> 00:55:53.679
But again, it's if you're not going to get it done until next December and it costs this much money, is it worth it for me?
00:55:53.760 --> 00:55:54.000
No.
00:55:54.159 --> 00:55:58.639
If you can get it done by next Friday and it costs this much money, I don't care if it takes you five minutes.
00:55:59.199 --> 00:56:03.760
Because to me, it's worth this much money to be done next Friday, period.
00:56:04.800 --> 00:56:14.719
So it's it's it's it's uh it's a change in the professional services industry, and I think it's um slowly being adopted across many uh industries.
00:56:15.039 --> 00:56:15.760
Yeah, yeah.
00:56:15.920 --> 00:56:18.880
We're doing some interesting things now in that regard.
00:56:18.960 --> 00:56:22.960
It's it a lot of what we do is still time-based and should be.
00:56:23.280 --> 00:56:30.320
Knock on wood so far, we're we're seeing more opportunity to offset the reduction in labor.
00:56:30.400 --> 00:56:32.719
So we haven't been negatively impacted.
00:56:33.440 --> 00:56:50.000
But where we are seeing opportunity is in in uh where routine, mundane, doesn't requ it requires expertise from a software engineering perspective to execute, but it's not the whiteboard, white space, big strategic thinking and design work.
00:56:50.880 --> 00:56:51.519
Maintenance.
00:56:53.440 --> 00:56:56.559
The maintenance that needs to be done in in applications.
00:56:56.639 --> 00:57:03.119
We we now have uh a product that's subscription based that the engineers aren't doing the dishes.
00:57:04.000 --> 00:57:04.239
Right?
00:57:04.559 --> 00:57:19.360
You know, we can we can monitor the bill of materials, find the CVEs, the agentic workflows, handle C VE remediation, automates that all the way up to the pull request, notifies the engineers so the human in the loop is still there from a quality assurance perspective.
00:57:20.400 --> 00:57:25.679
But it's also generating deep uh technical documentation with every commit.
00:57:26.639 --> 00:57:43.920
And so if you think about, you know, as an as an engineering team, it might take if we take on somebody else's software application, we might be looking at 80 to 100 engineering hours to create the level of documentation we need because nobody's good at documenting their custom software applications, it just doesn't happen.
00:57:45.280 --> 00:57:47.599
You can do that in minutes as part of this platform now.
00:57:47.679 --> 00:57:50.639
So I've so that one is value-based priced.
00:57:51.119 --> 00:57:52.159
Yeah, no, I agree with you.
00:57:52.239 --> 00:57:54.880
And that's how I've seen better documentation.
00:57:55.039 --> 00:57:56.320
Again, you can overdo it.
00:57:56.480 --> 00:57:57.119
Yeah, right.
00:57:57.280 --> 00:58:00.400
I mean, AI can create some great things, but you still need to work with it.
00:58:00.559 --> 00:58:20.639
I keep saying I've seen better documentation than I've ever seen in my life coming from developers with modifications because the AI tools will create even simple things of these are the changes that I made, these are if you're dealing with tables, these are the pages, forms, or UI, this is what was added and such, and this is how this works, right?
00:58:20.719 --> 00:58:24.000
You still can guide it, it just puts it together so that somebody can follow it and use it.
00:58:24.079 --> 00:58:35.920
And if somebody can also come in and pick up um to understand the application if they need to maintain it, yes, and also if you have AI agents working with you, that documentation can be used for those AI agents to help you maintain it, which is key.
00:58:36.079 --> 00:58:37.760
But you did say something which is also important.
00:58:37.840 --> 00:58:43.360
Again, I want to emphasize before everyone starts telling me because we've had these conversations about value versus time.
00:58:44.400 --> 00:58:46.000
Uh we all said it.
00:58:46.239 --> 00:59:08.239
All of us, yeah, uh there are points for time and materials, yes, and time and materials are necessary for some things, other things are more appropriate for uh value-based things, and neither one of them are in a position to um uh or take advantage of you know whomever's consuming those.
00:59:08.400 --> 00:59:14.880
It's the appropriate billing cycle or method for the bill of, in my opinion.
00:59:15.039 --> 00:59:19.599
And I I think that's uh a different way to frame looking at it.
00:59:20.400 --> 00:59:33.199
Um, because uh if you do all of one, then you could uh at this point impact um the revenue that you're making for a professional services industry.
00:59:33.440 --> 00:59:34.000
Yep.
00:59:34.480 --> 00:59:39.440
And uh again, that's many industries, not just the tech industry.
00:59:39.760 --> 00:59:40.239
Certainly.
00:59:40.400 --> 00:59:42.480
Uh so Mr.
00:59:42.639 --> 00:59:44.880
Matt, I could talk to you for ages.
00:59:45.039 --> 00:59:55.039
I really wish we did get to dive into how some of uh you're using AI to uh streamline your business uh and uh you know some of the tools that you've adopted.
00:59:55.199 --> 00:59:58.320
But maybe we'll have to have you back on again to jump into that.
00:59:58.559 --> 01:00:14.480
But uh uh I wanted to get your intake on how or your excuse me, your input or your take on how to get started with all of this because that seems to be the bubble that I'm finding in as I get out of the AI bubble and into the business bubble.
01:00:14.639 --> 01:00:16.559
And uh there's far more people concerned about that.
01:00:16.880 --> 01:00:17.679
100% agree.
01:00:17.760 --> 01:00:19.440
Yeah, it was a pleasure being here, guys.
01:00:19.519 --> 01:00:20.400
I really appreciate it.
01:00:20.559 --> 01:00:21.280
Lots of fun.
01:00:21.519 --> 01:00:24.400
Always have a good time talking with you guys, really do.
01:00:24.480 --> 01:00:26.960
So uh look forward to being back at some point and hopefully.
01:00:27.280 --> 01:00:27.519
Absolutely.
01:00:27.679 --> 01:00:28.880
We'll get you on in a few months again.
01:00:29.840 --> 01:00:32.880
No, absolutely, and I'll have to have my soccer season will still be on.
01:00:32.960 --> 01:00:40.880
We'll have you on before the soccer season ends, and maybe we'll have to do it when uh I'll take a look at the schedule, see when Revolution plays Cincinnati, and we can do it the day after.
01:00:41.199 --> 01:00:41.840
So we can talk about that.
01:00:42.000 --> 01:00:42.719
That would be great.
01:00:42.880 --> 01:00:44.559
I would love to hopefully we'll win.
01:00:44.639 --> 01:00:45.599
Uh so we'll see.
01:00:45.760 --> 01:00:57.599
But uh, if anyone would like to reach out and contact you to talk a little bit more about AI, how they may be able to adopt AI within their organization, or see how you're using AI in your organization or any of the other services that you offer.
01:00:57.760 --> 01:00:59.119
What's the best way to get in contact with you?
01:00:59.519 --> 01:01:00.320
Two great ways.
01:01:00.480 --> 01:01:02.960
Look me up on LinkedIn, Matt Striplehoff.
01:01:03.039 --> 01:01:05.760
That's all you gotta do is search for and uh send me a message.
01:01:05.840 --> 01:01:10.639
Uh alternatively, you can just go to redhawk-tech.com, fill out a contact form.
01:01:10.719 --> 01:01:14.000
I'm the one who gets uh that information routed to me through the marketing team.
01:01:14.159 --> 01:01:17.599
I'll be happy to schedule a call and have a conversation.
01:01:17.920 --> 01:01:18.239
Great.
01:01:18.320 --> 01:01:18.719
Thank you.
01:01:18.800 --> 01:01:22.800
I really appreciate you taking the time to speak with us again today, and I look forward to speaking with you soon.
01:01:22.880 --> 01:01:23.280
Cha-cha.
01:01:23.840 --> 01:01:24.559
Thanks, Matt.
01:01:26.480 --> 01:01:31.599
Thank you, Chris, for your time for another episode of In the Dynamics Corner Chair.
01:01:31.760 --> 01:01:33.760
And thank you to our guests for participating.
01:01:34.000 --> 01:01:35.599
Thank you, Brad, for your time.
01:01:35.760 --> 01:01:39.280
It is a wonderful episode of Dynamics Corner Chair.
01:01:39.440 --> 01:01:42.800
I would also like to thank our guests for joining us.
01:01:42.960 --> 01:01:45.760
Thank you for all of our listeners tuning in as well.
01:01:46.000 --> 01:01:49.840
You can find Brad at developerlife.com.
01:01:50.000 --> 01:01:54.480
That is D V L P R L I F E dot com.
01:01:54.639 --> 01:02:00.159
And you can interact with them via Twitter, D V L P R L I F E.
01:02:00.800 --> 01:02:09.039
You can also find me at mattalino.io, m-a-t a l i no dot io.
01:02:09.840 --> 01:02:13.519
And my Twitter handle is mattalino16.
01:02:14.320 --> 01:02:17.280
And see you can see those links down below in the show notes.
01:02:17.440 --> 01:02:18.639
Again, thank you everyone.
01:02:18.800 --> 01:02:20.559
Thank you, and take care.