Subscribe
Share
Share
Embed
The Value Creation Mindset explores the decisions successful leaders make to create real, lasting value for their customers, teams, and businesses.
Hosted by A.J. Singh, each episode features candid conversations with founders, CEOs, and builders who have been in the trenches, unpacking hard-earned lessons on leadership, technology, strategy, and execution. No hype. No shortcuts. Just clear thinking, first principles, and practical insight from people who have actually built things that work.
00:00:06:01 - 00:00:40:01
Unknown
Hi there. Welcome to the Value Creation Mindset podcast on your host RJ Singh. This show is about the the choices successful business leaders make to create maximum value for their customers and and themselves. Today I am joined by Peter Van Schlick, CEO of XM Pro. Peter, I'm sure I bought your name. What's the right pronunciation?
00:00:40:03 - 00:00:57:19
Unknown
All right. Very good. And where? Where are you? As of right this moment, Peter.
00:00:57:21 - 00:01:52:15
Unknown
Very nice, very nice. Yeah. I'm in. Not so sunny Seattle, at the moment. All right, so in this episode, we will unpack the decisions that change the the the trajectory of Peter's business. Talk a little bit about the trade offs made, what you would do again, what you wouldn't do again. And basically try to teach our listeners and our and our and our viewers the frame of mind you found that creates the most value for your customers and for yourself and for your business is, at the end of the day, the basic premise of this podcast and and my and and my perspective is that any business leader always starts with the same
00:01:52:19 - 00:02:40:11
Unknown
inputs. You've got people, you've got technology, you and you have some amount of points and a jobs to to focus all of those inputs through a lens of shared vision and, and hopefully effective leadership in order to create maximum value. And it's the it's that focusing effect that, that, is so hard to find because I think we've all seen cases where a company and a leader has a lot of capital, excellent people, and amazing technology, and they still end up lighting it all on fire.
00:02:40:13 - 00:03:09:22
Unknown
So, the choices that you made that the that you fall, that you avoided are what I'd like to explore a little bit more on this call. On this podcast, I should say so that said, Peter, you're a mechanical engineer by training, as am I. So that we have kind of a common, common ground there.
00:03:09:24 - 00:05:16:02
Unknown
How did your experience and training as a mechanical engineer shape your outlook and shape the decision making process, which you've embraced over.
00:05:16:05 - 00:05:41:22
Unknown
No, I, I love that. I, I remember when I was in college I would talk to, to my father who was also a mechanical engineer and in every single call he would tell me two things. One, focus on the fundamentals and you can derive everything else. So it made me realize there are fundamental first principles. You're studying mechanical engineering.
00:05:41:22 - 00:06:01:26
Unknown
You get to believe it when you shift in the software. You. I mean, I believe that because I came from the, the mechanical world, but a lot of computer science folks, a lot of people have just been coding for all their lives. They still don't. They still haven't quite grasp that there are timeless first principles there as well.
00:06:01:28 - 00:06:49:05
Unknown
Some do, some don't, I suppose. But the second thing that my father always told me, and this is how he ended every single phone call. He ended up telling me, stay rational. And in my life, as I've grown the business that still rings in my ear. So I certainly got that.
00:06:49:08 - 00:07:29:29
Unknown
Right. Exactly.
00:07:30:02 - 00:08:06:09
Unknown
You know, it's it's there seems to be a a rush for software folks to. You have an idea. You you want to jump from the whiteboard to the keyboard. And some of this is, is. Part of the the the the excitement and the software space has always been. If you can come up with an idea, the solution to a problem, which for me always comes in the shower or in the bathroom generally I don't know why, but that that and fairly common phenomena.
00:08:06:11 - 00:08:31:13
Unknown
You can work on a solution and you can bring it to life, and you can see your results just with your own two hands or with the hands viewed a couple of t numbers, but at the same time, you know, I find myself asking my team members and my clients the same thing also. I mean, you know, have you design something?
00:08:31:16 - 00:09:17:14
Unknown
Are you going from the whiteboard to the keyboard? I thought it was a good idea to design something before you build it. I know it's a novel idea, but I would think in your space, in the oil and gas space and heavy industry, I mean, when you've got contact with the physical world, with physical machines, I mean, it's it's similar to, I think my, my experience in the IoT space when you add a hardware component to it, all of a sudden things become a lot more real and less esoteric because a decision you can help a customer make, has a massive physical impact on a job site, a company, the profitability of a
00:09:17:14 - 00:12:22:08
Unknown
particular project, success or failure of the overall endeavor. I mean, in the IoT space, if you screw up one firmware build and you push it out over the air to 10,000 devices, you think you can afford to make 10,000 field visits? Not so much. But that's just, just kind of I do you find that that running close to the metal or close to the physical world really helps to shape and drive your software teams to be a little more pragmatic and rigorous.
00:12:22:10 - 00:12:32:12
Unknown
Yes.
00:12:32:14 - 00:13:46:03
Unknown
Okay.
00:13:46:06 - 00:14:16:08
Unknown
It is really hard. You got to calibrate, you know, where you invest, where you apply pressure. So. So you actually deliver value versus, again, I mean, I've had I've had clients call me, ask me a piece of advice, five minute conversation. And so many times they've said, God, I wish I called you six months ago. You would have saved me six months of team effort if.
00:14:16:10 - 00:14:41:27
Unknown
And I'm like, sometimes you need that level of guidance and just basic core strategy that we with our gray hair, we we have a bit of a, a bit of perspective. And I think for a lot of, a lot of team leaders, a lot of, a lot of engineering teams, they're very they're two inches away from it.
00:14:42:00 - 00:15:22:03
Unknown
That's all they see. And it's very easy to waste years working on something. And when you ask the hard questions of, okay, well what value are you really creating here? I, I had, a case where a friend of mine called me in to, do what we call a tech 360, which is just an overall review of R&D operations from a business perspective, from a core asset purchase perspective, review the portfolio, the tech stacks, processes, team leaders, team members, everything.
00:15:22:03 - 00:15:51:03
Unknown
So we we invested a lot of time there. We came up with it. We came up with our assessment, when that would the CEO, my my my friend, laid it all out and his conclusion was the same one that that we kind of guided him through. So he looked at and said, so it looks like I've lit $20 million on fire over the last couple years.
00:15:51:05 - 00:16:24:07
Unknown
And I'm like, yeah, I accessed the reality of it. And he was appreciative because we provided him a plan to turn things around, cut, cut costs in half, and, dramatically increase value creation. Right. And, so it was a clear path forward, better for everyone. And I did my best to make this difficult news. As easy to hear as possible.
00:16:24:10 - 00:16:56:29
Unknown
It wasn't it wasn't good enough because I got a call, a few days later after my friend, the, now former CEO, explained the situation to his boss, who was the head of the a conglomerate they were a part of. And he fired them on the spot. So I got my friend fired for telling him the company the truth.
00:16:57:02 - 00:17:27:29
Unknown
In a way that I made as hirable as I could. But these are some of the very difficult, hard to hear truths that that determine whether you create value or not. And if you don't, if you don't kind of touch grass and say it asked continuously, all right. Is what I'm doing actually going to have a, return is at some level depending on the org, depending on the organization.
00:17:27:29 - 00:17:59:05
Unknown
Obviously. I mean, sometimes technical leaders are more infatuated with particular technologies or building a fiefdom rather than solving the problem and delivering value and, and, and cap, what is wrong with us? I mean, I, I don't is is it just I mean, do you do you sometimes feel like the old man telling kids to stay off the grass?
00:17:59:05 - 00:18:52:09
Unknown
Because I do sometimes.
00:18:52:11 - 00:20:19:09
Unknown
The.
00:20:19:11 - 00:20:35:01
Unknown
Right.
00:20:35:03 - 00:21:05:17
Unknown
Right, right.
00:21:05:19 - 00:21:38:28
Unknown
No, I, I love that, you know, it it shows a maturity in that, in that business when you know, there are these, these four core buckets. And, I mean, you know, you work at anything long enough, and you sift through the noise long enough, you hopefully lock on to the signal, the the handful of timeless things and principles and pillars that that really need to guide every single decision.
00:21:39:01 - 00:22:21:21
Unknown
So in the, in all the looking in the mirror that I've done over the years, there are there are again and it's weird, there always seem to be four. I found four core pillars or buckets in commercial software to, to development. Every decision, everything you possibly do, every engineering hour that is invested should either have a positive impact and revenue retention, net new revenue generation, increasing lifetime total value for for your customers, or cost reduction.
00:22:21:23 - 00:23:01:06
Unknown
And this doesn't matter if you're paying for tokens or if you're investing in a new platform or buying a new machine or tripling down on agile, which my god, to me, please stop this already. I you know, it's like, yeah, you you haven't figured out that agile generally equals make it up as you go along. But I mean, every decision, even down to what do you name a variable in a class or a method that no one ever is going to see for another 15, 20 years, except for the person who has to maintain whatever you might be building.
00:23:01:08 - 00:23:31:02
Unknown
Even that can be categorized in these buckets. So institutionalizing that, just just as you said, in mining these buckets up and down, the organization, everybody understands them. It is it is part of their shared vision. And they know that if you want to move the needle or get anything approved, it has to be put in that context. I, I'd love to see that same thing happen in the software space.
00:23:31:04 - 00:23:56:21
Unknown
More more generally. And, and, what can we do other than reinforce these fundamentals with whoever we engage with? Because they are fundamentals. All right. Aren't they?
00:23:56:23 - 00:24:53:12
Unknown
Sure.
00:24:53:14 - 00:25:00:21
Unknown
Chris.
00:25:00:23 - 00:25:38:28
Unknown
You know,
00:25:39:01 - 00:25:53:17
Unknown
Yeah.
00:25:53:20 - 00:26:00:02
Unknown
If I'm wrong.
00:26:00:04 - 00:26:42:12
Unknown
No, no, I, I you couldn't be more right? Right. The foundation of a platform, a product, I mean, let's make it. There isn't enough of a delineation between software development and product development. Software development? Sure. Five code. It code an hour, everyone. It's just coding. Look, my nine year old daughter can code and and with, Claude, my my 11 month old puppy can probably code.
00:26:42:14 - 00:27:22:19
Unknown
That's not the freaking point. What's the context? Are you. Are you laying down a very solid and completely deterministic foundation, an architecture on which you can leverage Lims, other inference technologies to create even more, more, more a value. But if your foundation isn't solid and timeless, if your foundation, your your core architecture and it's it's not even so much about particular language or frameworks you end up using, it's the header and what thoughts have been put into it.
00:27:22:21 - 00:27:54:02
Unknown
You know, it's it's like if your foundation is unstable, you can't build anything on it, you can't sell it. You can generate a single dollar off of it if it's if it's stable, but it's not scalable, forget it. What was the point? You can't grow if it's stable and it's scalable, but it's not profitable because you're consuming too many tokens to run it or or you built everything in AWS Lambda without checking the apex and, and any of this stuff.
00:27:54:04 - 00:28:23:09
Unknown
What? Why even start if it's not profitable? And once it's stable and it's scalable and it's profitable, the fourth requirement, if you want long term success on this, it has to be serviceable. But what does that mean. And it's well, novel idea. You don't build a throwaway product. It's not building software. You're building a product which happens to be software.
00:28:23:12 - 00:28:57:24
Unknown
But to make something serviceable inherently means you've thought about how to service it. You've made choices architecturally from a tech stack standpoint, from a development methodology perspective, from a back end schema, first, first, first person perspective. You've thought about it and designed it and architected it and structured it in as modular of a way as you can. So you can adapt, you can build on top of it.
00:28:57:24 - 00:29:28:16
Unknown
You actually have a solid foundation and you're not built on sand. But as you said, I mean, again, I living in the real world with real industries, with with machines, of any kind. I it just puts much more focus on those and those timeless fundamentals that a lot of, a lot of people don't seem to stress often enough.
00:29:28:18 - 00:30:14:22
Unknown
And in, in, in the great new world of generative AI, the reality is to get real value from it. 80 to 90% of the effort is getting all the structure and the plumbing correct around it.
00:30:14:24 - 00:32:27:02
Unknown
Oh my God.
00:32:27:04 - 00:33:07:03
Unknown
It's.
00:33:07:06 - 00:33:42:29
Unknown
The.
00:33:43:01 - 00:34:03:13
Unknown
Pure.
00:34:03:15 - 00:34:13:29
Unknown
Exactly.
00:34:14:01 - 00:34:27:14
Unknown
Yes.
00:34:27:17 - 00:35:08:20
Unknown
You're you are so exactly right. I, I, I just read an article, about the, former co co-founder of open AI, just, stating that the time of scaling is over. Now it's time to go back to research, because there are some fundamental limitations in lens as as people continue to chase AGI, which. Okay, fine.
00:35:08:23 - 00:35:43:23
Unknown
You've got money. Go do that. I'm not a galaxy brain. I'm not I'm not quite so interested in in in doing that. But the most interesting point in the article was a refound embrace of, innate constraints, meaning business logic, meaning fundamental code that understands and implements rules for a particular problem space which are definitional. These are structural.
00:35:43:23 - 00:36:23:07
Unknown
Is your foundational these these are laws because no other Lem is aware of laws of physics or or laws of mining or anything else. But I it's it's the. So I guess even the galaxy brains are coming to this conclusion, which is nice to hear. Hopefully the market correction isn't isn't too bad.
00:36:23:09 - 00:37:53:09
Unknown
For.
00:37:53:12 - 00:38:25:24
Unknown
Sure. No, no, it nor look, you know, it's it comes down to a fundamental question. Do you need to know the fundamentals? Do you need to know the problem space. Do you need to know the laws of physics? Why when you can ask, and it'll tell you fine. Even if you learn it from I do say at best, what's it going to be?
00:38:25:24 - 00:38:58:26
Unknown
93% accurate. Are you going to believe everything you know? It's it's we've been in the generative in the code generation space for 26 years now. So, November 18th was our 26 anniversary. And at this point we've got the ability to generate about 80% of the total code base needed for commercial platforms and products in a 100% deterministic way.
00:38:59:00 - 00:39:46:08
Unknown
So you push the button 100 times, you get exactly the same results. And now we are folding in, Gen I components, components to build on top of that deterministic foundation, which is again built on timeless principles. Right? I mean, it's it's not that it's not that difficult of a concept to embrace, but, when, when. Customers are thinking that all they need to do is ask ChatGPT what they should do or but, you know, you have to talk them down sometimes, don't you?
00:39:46:08 - 00:40:25:17
Unknown
Just in terms of managing your expectations and I've, I've read articles. So just saying that the number of board members can be reduced by about 20%, because now you can have AI chat bots replace them. And I and I suppose if you're that easily replaceable with the chatbot, yeah, maybe you should go in. Maybe you're not adding as a board member, but the, the the additive way in which you've integrated genocide and, and LMS makes a ton of sense.
00:40:25:19 - 00:42:28:25
Unknown
How well is that working for you in reality? And, how widely have you deployed it? Is it creating that net incremental value you were hoping for?
00:42:28:27 - 00:42:53:16
Unknown
For.
00:42:53:18 - 00:44:18:27
Unknown
The.
00:44:18:29 - 00:45:37:09
Unknown
Bot.
00:45:37:11 - 00:45:47:25
Unknown
The.
00:45:47:27 - 00:46:21:05
Unknown
Absolutely.
00:46:21:08 - 00:46:59:26
Unknown
Right.
00:46:59:29 - 00:47:40:23
Unknown
And our wives remind us that processes. We really suck. Yeah. No, I I'm I I get it I, I that particular task like lms and machine like I generally excels at pattern recognition. And if you've got a large continuously updated data set which you need to filter and in order to identify which alarm, say, or element and which might not be, but that's an ideal case.
00:47:40:26 - 00:48:19:24
Unknown
I mean, in, in, one of our IoT clients has a fleet of half a million devices. These are, dashcams in, in, trucks, and they are continuously monitoring, driving behavior, through the true AI deployed at the edge and processed and further filtered in, in the back end platform and then ultimately pushed out to a set of traps.
00:48:19:25 - 00:48:58:00
Unknown
But when you're watching, say, when you've got a driver facing camera and this drivers covering up, thousands of miles, the AI is smart enough to detect, right. Is the seatbelt connected or not? Is the driver smoking is the driver, drinking? Is the driver sleepy? Is the driver not looking right in front of them? Right. And sure, you can generate alerts and alarms for all of these actions.
00:48:58:02 - 00:49:31:00
Unknown
And just because you can doesn't mean you should. How many of those actually create any value in terms of risk mitigation for the fleet operator. So all they really care about is risk. And I mean, that drives insurance rates and everything else. So decreasing the noise and separating the signal from the noise, I think is an ideal. It is an ideal use case for alarms, as long as again the final determination is made by the human, not in every case.
00:49:31:04 - 00:50:00:00
Unknown
But you want to leave that open because again, if you've only got a night, if if you've got an agent that can tell you that there's 93% probability that your house is on fire, that's data you want to know, or that your house is about to catch on fire. Just just to push this a little bit further or that your extraction rig is about to fracture or have an event.
00:50:00:02 - 00:50:35:05
Unknown
This is all creating massive value for the customers and the the that the, the. What's been the biggest, what keeps you up at night though, about how much customers are willing to turn over in terms of decision making? Because obviously you have a great deal of influence on how you advise them, how you package the overall overall solution.
00:50:35:07 - 00:52:17:25
Unknown
How far do you want to push this in in terms of autonomous decision making and what kind of safeguards do you think you would, you would, you would want to put in or you've already put in. You spent a lot of time thinking about.
00:52:17:27 - 00:53:27:12
Unknown
Sure.
00:53:27:14 - 00:54:52:23
Unknown
Right, right.
00:54:52:26 - 00:56:05:14
Unknown
Right.
00:56:05:16 - 00:56:26:14
Unknown
Didn't they if you get in the way back machine, didn't they used to call these expert systems back in the day?
00:56:26:16 - 00:57:05:26
Unknown
Right on the.
00:57:05:28 - 00:57:22:24
Unknown
It's the the the term that that that was ringing in my brain as you were describing. This is you need guarded actions. You know, they'd have to be guarded by known.
00:57:22:26 - 00:57:53:23
Unknown
Deterministic rules. And, the getting the mix right is an interesting challenge. And and no doubt, there'll be new innovations that that are coming up shortly. You know, the, the. Where where are you thinking? I mean, where do you think things are going to go next? What's the.
00:57:53:25 - 00:59:13:13
Unknown
What are you most excited about that you're that you'll be working on over the next 6 to 12 months?
00:59:13:16 - 00:59:32:14
Unknown
Right.
00:59:32:16 - 00:59:40:05
Unknown
But.
00:59:40:08 - 01:00:05:14
Unknown
I love it. I, you know, the the when you said earlier you need to know what failure looks like. That's. I found that very interesting because oftentimes, you know, at the start of an engagement or when you're, when you're getting into something, you have to ask the question, what does success look like? So we all know what to look towards.
01:00:05:16 - 01:00:33:18
Unknown
But it's just as important to ask what does failure look like now? Many cases the answer if people are going to be truthful is, well, it looks like what we're at right now, but in this case, when you're dealing with physical AI automation at scale, at that level, it's a whole other level of safety that you need to achieve.
01:00:33:20 - 01:00:43:14
Unknown
I it sounds like an awesome challenge. Honestly.
01:00:43:17 - 01:02:06:19
Unknown
Like.
01:02:06:21 - 01:02:09:01
Unknown
Yeah.
01:02:09:03 - 01:02:42:06
Unknown
We do, don't we? Yeah. I, in the early days when we were looking for money, we, you know, went to a potential investor. We told them what we were trying to do, which is fundamentally automate, amass a massive portion of the commercial software product instruction process. And he looked at me and he said, AJ, why are you trying to do something so hard?
01:02:42:09 - 01:03:06:07
Unknown
And then he he pointed to, another company that he invested in. This is in the early to this early 2000, said, look at these guys. Two guys working out of a garage selling cell phone ringtones at $0.99 apiece, and they're already at $2 million a year. So why are you trying to do something so hard? And I, I went home and I thought about that.
01:03:06:07 - 01:03:32:27
Unknown
It's like, you know, there are easier ways to make money, but none of them are quite as interesting or challenging as this. So 26 years later, I mean, I'm still at it, but I, I, I get the drive for that. And I'd like to close the day by asking you, I mean, you know.
01:03:33:00 - 01:04:08:10
Unknown
There's always a reason that we get up in the morning and, For, for me, I really believe our legacy will be measured by how well and how far we lift all those we touch. And there are lots of ways of creating value for our customers, their customers, their employees, our employees, ourselves, everyone who we who we touched.
01:04:08:10 - 01:05:36:08
Unknown
But what if what do you want your legacy to be?
01:05:36:11 - 01:07:08:27
Unknown
Right.
01:07:08:29 - 01:07:31:25
Unknown
I, I get you. I appreciate you, Peter. This is this is, Oh, you get to a certain point in life, and and you ask yourself, why am I doing this? And we all need a purpose. And I.
01:07:31:27 - 01:08:04:11
Unknown
I respect yours, I. Thanks for going at it. I mean, you know, it's it's. You you learn we always have more choice than we think we do. And I went through life for a long time thinking, man, I wish I had have more choices. And in hindsight, I did. I just didn't realize it. So, thanks for your time.
01:08:04:13 - 01:08:18:15
Unknown
Thanks for coming on. And we're on our podcast and sharing your insights. I, wish you all the best. I have, have no doubt. You'll move the needle.