How data turned $120M into $1.8B | 28X Private equity exit with CIO Michael Franklin (Ep 004)

In this episode of The DATA SECRETS Podcast, I talk with Michael Franklin, Founder and Managing Partner at Cold Iron Capital and a multi-exit Private Equity CIO, about how data and technology can transform even the most traditional manufacturing businesses.

Michael shares his journey from tech leader to business operator, detailing how he helped grow cabinet manufacturer ACPI from a struggling $120M company to a $1.8B powerhouse with a staggering 28x return for its PE backer. We dig into the critical role of operational data, the challenges of merging legacy systems during rapid M&A, and how Michael’s “customer-first, data-backed” approach surfaced hidden margin leaks and fueled exponential growth.

We discuss:
• Why blending deep IT and business management experience is a game-changer for aspiring CIOs
• How to use “on-time complete” as a north star metric for operational excellence and customer satisfaction
• Building practical, shop-floor-ready dashboards from legacy ERP (AS400) data…no fancy tech required!
• Tackling the chaos of multi-ERP sprawl during acquisitions, and the real-world playbook for BI consolidation
• The critical moment data revealed that “biggest” wasn’t “best” in customer profitability, and how that changed strategy
• Avoiding expensive data horror stories by investing in process, clarity, and business alignment (not just tech)

Michael also unpacks the evolving opportunity at the intersection of manufacturing, private equity, AI, and cybersecurity, and why now is the perfect time for smaller, data-savvy firms to leapfrog their competition.

If you’re a CIO-in-waiting, PE-backed operator, or just want to hear how relentless focus on data and process can drive world-class outcomes, you won’t want to miss this episode.

Connect with Michael: 
🔗 LinkedIn: https://www.linkedin.com/in/michaelfranklin/ 
📧 Email: mike@coldironcapital.com

Follow the DATA SECRETS Podcast

📬 Get episode recaps & bonus insights at allegroanalytics.com/podcast
🕺🏻 Connect with Nathan on LinkedIn
📺 Watch every episode on our YouTube Channel

Chapters

• 0:00 From Tech Leader to Business Manager
• 8:34 Cross-Department Experience for CIO Success
• 12:52 Grateful for Gaining Experience Before MBA
• 21:01 Legacy Systems Enable KPI Tracking
• 23:38 Legacy System Upgraded with Data Warehouse
• 28:40 Building Effective Data Systems
• 33:28 Decoding ERP for Delivery Management
• 38:19 ACPI's Transformative Merger of Equals
• 43:31 Rethinking Customer Value Assumptions
• 49:00 Catalyst for BPM System Development
• 56:05 Challenges in Operationalizing AI Tools
• 1:02:50 Just One Business Tool
• 1:04:17 Embrace Upskilling for Business Success

Transcript

Michael Franklin [00:00:00]:
I went to a firm called ACPI, which is a cabinet manufacturer. And when I joined, it was, it was actually losing money, you know, and it was that within three months we got it to break even. Over eight years we grew it from, I think it was like 120 million in revenue when we started and we exited at 1.8 billion revenue.

Nathan Settembrini [00:00:20]:
Your data has secrets, secrets that could change everything if you only knew where to look. Welcome to the Data Secrets podcast. Welcome to the Data Secrets podcast where we uncover how business leaders crack the code, find truth in their data and turn insight into action. I'm your host Nathan Settembrini and today we have Michael Franklin. Michael is a multiple time CIO with multiple exits in private equity. Now turned investor. He's the founder and managing partner at Cold Iron Capital and he's so confident in technology as an accelerator for business that he's buying businesses and driving performance with data and technology. Mike, welcome to the podcast.

Michael Franklin [00:01:10]:
Yeah, thanks Nathan. Happy to be here.

Nathan Settembrini [00:01:12]:
Awesome. Awesome. So, for the, for the audience, you know, I imagine some people might know a bit of your story, but most people probably don't and so could you just introduce yourself a little bit more kind of what you're doing now and also how you got here?

Michael Franklin [00:01:26]:
Yeah, yeah, no, happy to do so. So for about the past 15 or years or so, really been focusing in the private equity space, working primarily on turnarounds in a, like a CIO, CTO type of capacity. You know, the PE world is exciting. It's, you know, stakes are high, timelines are high, reward is potentially high as well. So that's been a good run. Had a lot, learned a lot, had a lot of fun there, met a lot of great people and recently transitioning out of that CIO, those full time CIO roles into more venture investing, doing a little bit of mentoring of some kind of up and coming tech leaders as well as like you mentioned, I'm strong believer in technology and data's ability to help companies grow. And to that end myself and some friends are, you know, sort of looking to start some ventures or buy some acquisition, buy some businesses and help them grow through technology and data. So kind of how, how I got here, you know, probably not, not exactly a straight line. I'm sure a lot of people could say that, you know, it, you know, people have far more interesting stories than a lot of them realize and I'm, I'm no different than that. So early in the career, spend a lot of time, time in the trenches, startups doing traditional technology type of roles, some development, infrastructure, those kind of things. Had some decent success there. You know, had a lot of great opportunities, a lot of people that believed in me, a lot of people that helped kind of pull me up and that culminated into an opportunity to pivot out of technology and into a pure business management role. So assumed sort of a couple different assignments over the course of several years at a company called American Gypsum. And you know, those, I will always be thankful for those roles and those opportunities that they provided. You know, tons of tons of great lessons learned, right? You, you think you understand something until it's your responsibility, then it's, then it becomes real, right? But you know, I sort of woke up each, each Monday morning just dreading going to work, you know, in this, this more general management type of role. You know, it was, it was very humbling going from like rock star tech leader to man. I'm not even sure I, this, I'm qualified to be in this room or in this meeting. So I had several years of really fighting that imposter syndrome and you know, was a tough couple years. I learned a lot and one of the most important things I think there was, hey, I, I, I picked up a bunch of knowledge that's just great, great, great for an IT leader, great for a CIO to understand what the business itself is trying to accomplish in a functional area. But you know, I definitely realized this is, this is not what I want to do for the next 20 plus years, right? So one day, just literally, honestly, I was probably trying to avoid answering emails, you know, or something like that, and scrolling through LinkedIn and came across a role for a CIO role at a private equity backed manufacturing firm. Turnaround situation, high risk, high reward, and you know, something about it. I'm like, hey, I think there's something there. Met the team that was involved in it on both the operator side as well as the PE side and immediately, you know, the spark was there. I was like, oh hell yeah. This is, this is what it's supposed to be like, right? Like, you know, everyone was super hungry. Everyone was open to any idea wanted, you know, like when it just focused on winning, right? And, and big ideas, big plans, and I'm very competitive. So that, that really, you know, sort of turned my lights, you know, kind of put the bright brights back on and, and reinvigorated a little bit. And I immediately pivoted over there. Sort of thanked my, my, my bosses at American Gypsum saying, hey, I really appreciate everything you did and, and went hard into the PE space and, and kind of haven't looked back since. With that, you know, I went to a firm called ACPI, which is a cabinet manufacturer. And when, when I joined, it was, it was actually losing money, you know, and, and it was that within three months we got it to break even. Over eight years. We grew it from, I think it was like 120 million in revenue when we started and we exited at 1.8 billion revenue. So like a hell of a journey, you know, and a lot of acquisitions and a lot of organic growth and we're able, you know, lucky enough and worked hard. We had a little combination of, you know, work hard, the right mix of people and some luck. Right. Like I'll be the first to admit, like there's more to luck in success than people probably want to want to admit to. But we ended up having a 28x return for the PE sponsor, which ended up being the largest in their history. So it was, it was a hell of a return, hell of a chance. We learned a ton. I learned a ton. I met some amazing people at both ACPI and then later Cabinet Works as well as, you know, the, the folks, the men and women inside private equity firm that was the owner. Amazing people, incredible intellectual horsepower. Like I think I do okay. These people were like off the chart above us, you know, really smart. So learned a lot of great things from them. So that's kind of what brought me to today is I've kind of, I liked that first bite at the apple and went back a few times and have served in similar type of CIO turnaround, type of capacities or digital transformation, kind of very similar and have sort of did that a couple more times and kind of brought me to today. Whereas, you know, I'm in fortunate enough position to be able to kind of close that chapter of full time CIO and pivot to doing something a little different going forward. Investing more venture investing. Like I said, some of my friends and I are very interested in buying some smaller businesses that we could really leverage with technology and kind of seeing where that goes. Right. Like, who knows? Who knows?

Nathan Settembrini [00:07:19]:
That's awesome. Yeah. So just to kind of focus on that transition from kind of general management into CIO level leadership. What was that transition? Like what. And if someone's listening, who potentially wants to plot their own path to be a CIO, like what. What tips or advice would you give for them?

Michael Franklin [00:07:41]:
Awesome question. So looking, I've, I've thought about this frequently. I, I've been asked frequently, you know, by people like, hey, how do I, how do I get to a CIO chair? And there's a bunch of paths. Obviously you have to command some deep knowledge in a technical area and then have to have a pretty good understanding across the other major areas in technology. But I think more than anything that really would set a candidate apart is some time in service in some type of business role. And this is often like, if you look at a traditional CIO, you'll often see they come from a consulting world. You know, they'll come out of one of the big consulting. Just because it's those roles that give a CIO a ton of exposure to actually like being embedded in a business function for some period of time, like solving that business function's problems or solving customer problems. So I would say if at all possible, if, if someone has that ambition to go that, you know, VP of IT CIO, CTO route, especially more the CIO route, if you can, if you can convince somebody to give you an opportunity, give you assignment to run a run like a part of the actual business, even you know, customer service, you know, work in the marketing, working sales operations, rev operations, manufacturing operations. Find somewhere where you're like, hey, can I, can I get a one, you know, a six month assignment over here and really, really get embedded in it? Those opportunities provide so much insight and experience that when it does come time to interview or be considered for a CIO position, you're going to be talking the same exact language as the people that want to hire you. Right. So it's, you know that because oftentimes, as you know us technologists, we love our acronyms, we love all this very specialized jargon and we forget we're the only ones that care at all about that jargon. And the rest of the business is like, no, man, I'm just trying to get this problem solved for our customer. Right. So that, that exposure, if a person can get that exposure in some way, I think that that provides so much benefit to them down the road.

Nathan Settembrini [00:09:56]:
Yeah, yeah. That's awesome. And then so you got into this PE world and you know, you had this amazing exit, like 28x ROI. Like that's insane.

Michael Franklin [00:10:09]:
Yeah, it really was once in a lifetime kind of thing.

Nathan Settembrini [00:10:12]:
Yeah. Yeah, that's really awesome. You know, and we'll, I will unpack how you guys did that. I think that's probably going to be the meat of what we talk about.

Michael Franklin [00:10:22]:
Yeah.

Nathan Settembrini [00:10:23]:
And so that cabinet manufacturing, I imagine that, you know, that business, a lot of moving parts.

Michael Franklin [00:10:33]:
Literally.

Nathan Settembrini [00:10:33]:
Yeah, literally.

Michael Franklin [00:10:35]:
Like in our main Factory, the flagship factory. In any given time, there were about 4 to 500,000 parts that were like work in process that were inside the factory campus at any one time. So yeah, it was cabinet manufacturing. Like anything, cabinet manufacturing is far more complicated than people realize.

Nathan Settembrini [00:10:54]:
Right, yeah. And so then in your later roles, you were CIO of other manufacturing companies, is that right? Correct.

Michael Franklin [00:11:01]:
Yeah. So. So I've, I've had, you know, I started my time at American Gypsum, you know, kind of just kind of going back to manufacturing executive roles here. American Gypsum, manufacturer of drywall, most people know it as Sheetrock. You know, that's, that's like a brand name but. And then ACPI, which later became Cabinet Works with a, with a merger that we can talk about. And then a few sort of smaller consulting type of roles in, you know, aerospace and some other things here and there's. And then my most recent time was in a steel based manufacturing firm, manufacturer of steel deck and steel joists. Like highly engineered, very complex manufacturing situation. So yeah, I've spent a lot of time in manufacturing. By far the bulk of my experience is in manufacturing. That's. Honestly, I think it's extremely interesting. It's great to go out on a shop floor and watch people build an actual thing, you know, like point to like, hey, we did all that work today, right? Like, that's fun. That's really fun.

Nathan Settembrini [00:12:01]:
Yeah. Yeah. My first, my actually first, first role out of Georgia Tech was at a German, like a high end kitchen and bath manufacturer.

Michael Franklin [00:12:11]:
Okay.

Nathan Settembrini [00:12:12]:
And so they would make everything from toilets and sinks to furniture, bathroom cabinets, kitchen cabinets, that kind of stuff. Company called DURAVIT. And it was. So I did that after engineering and got four years under my belt before going back to business school and that. Seeing everything from the manufacturing side to the distribution after sales support, customer service, all of that. I had so much context to tie business school to way more than if I had maybe worked at a software company or something like that. Cause I got to see it and be around it is really cool.

Michael Franklin [00:12:52]:
Yeah, I think that's important. Right. I got an MBA as well. And I very much. I had sort of two options of, you know, when I graduated with my undergrad, it was, hey, do I go into the workforce and try to like learn stuff or do I just go back to school and get an mba? And I was, I'm very thankful for. Some people are like, oh man, you know, you need to go and get some experience in the real world for a little bit, you know, and you'll Get a heck of a lot out, more out of an MBA program. I'm so thankful that I took that advice. And to your point, with that context, it just makes so much more sense when you're going through those programs for sure.

Nathan Settembrini [00:13:27]:
And so I imagine some people are interested in this PE world. I know it's caught my attention the past five years. I'm curious, so did you end up kind of sticking with one PE firm and they just kind of moved you around or did you, does that experience with that amazing win, did that then allow you to go to another PE firm and say, hey, I'm your CIO. Put me in place. Put me in place.

Michael Franklin [00:13:52]:
Yeah, so it's. Yeah, exactly. You know, it's like once you become a proven commodity because not everybody enjoy like anything, not everybody enjoys working in certain environments. And a PE backed company is a certain environment, just like a startup is that it has, you know, very clear. I really enjoy it. It provides a lot of clarity and sense of urgency and you know, a willingness to win and willingness to try things, you know, so it's a very good environment for some people and for those people that think they have that interest. Yeah, once you kind of get your first proven stamp, then yeah, you kind of, you were often reused. Right. And I have primarily done work for a single private equity firm out of New York called American Industrial Partners. They're the largest in, largest PE firm in like an industrial slash manufacturing world. And you know, I think private equity sometimes gets a bad rap. There's a few like anything. There's a, there's a few people out there that, that are just, you know, whatever. They, they don't paint the industry in the best light. That's the vast minority that most of the people I've always, I've met npe, very intelligent, very driven. Good people though, you know, it doesn't mean they're not good people. They're trying to do big things. Just like in the. Excuse me, in the startup world you find the same, you find the same people in venture capital, you know, they're out there trying to make their dent. You know, the old Steve Jobs trying to make a dent in the universe. They're, they're, they're doing the same thing. They're just doing it with businesses that are already built, you know, and aip, American Industrial Partners, you know, they have a. I always thought it was a very, very catchy or very real tagline of they buy good businesses that are on that, having a bad day. Right. Good businesses on a bad Day.

Nathan Settembrini [00:15:38]:
I love that.

Michael Franklin [00:15:38]:
And. And. And that's. That. That's really what they do. You'll see on the press people, you know, again, news likes to badmouth anything because it sells, but PE firms really are the ones I've dealt with. They're very invested in the people, the companies, their communities those companies are found in. And they're. They're trying to do everything that they can to make sure those companies can win. They build them better. Because you got to remember, a lot of the investors that are behind companies like AIP or. Or BlackRock or, you know, larger ones. It's. It's retirement funds. It's, you know, California teachers retirement. It's New York pension. Right. So it's. It's, you know, they're investing on the behalf of a lot of Americans. It just. People don't really kind of put all those things together.

Nathan Settembrini [00:16:23]:
Yeah. Yeah. Cool. All right, well, let's. Let's start talking about some of these. I think we should focus our time on the. The ACPI cabinets, cabinet works, and so maybe dusting off some. Some cobwebs here because.

Michael Franklin [00:16:40]:
Yeah, yeah, it's going back a little ways, but no, we're good.

Nathan Settembrini [00:16:43]:
And so, you know, when I, When I start thinking about data analytics, like, trying to understand what's happening in the business, the most important thing is to kind of start with the business first. And so could you help our audience understand the cabinet business? You know, who are your customers? What's the business model? How did you guys differentiate? And then I do want to get into, like, when you stepped in. We'll get there in a second. But like.

Michael Franklin [00:17:14]:
Yeah, but just.

Nathan Settembrini [00:17:15]:
Just kind of looking at the cabinet business as a. As a whole.

Michael Franklin [00:17:20]:
So, you know, classic manufacturing, discrete manufacturing, which just means, you know, we're building a single thing out of a bunch of other discrete parts. So really, what made it different here? So manufacturing based out of one very large campus facility. And then there was multiple channels of distribution. So, you know, a traditional distribution channel, a dealer channel, which would sell more like repair and remodel. So say Mr. And Mrs. Smith wanted to remodel their kitchen. They would go to a local dealer in their area. That dealer, in turn would design the kitchen and then sort of pass along those custom cabinets to ACPI, the company I worked for. So that was a very traditional. And then there was also, like, a new construction. So either like a new. You go out and you buy a pulte home or something like that. That would be another avenue where we would sell product into and Then multifamily. So big high rises, universities, dorm rooms, that kind of things. So those are the main channels really, is that you have one that's distribution. You have one that goes for repair and remodel in the dealer community, and then, you know, more basic new construction, essentially. So that's. That. That's the sort of underlying structure of that business. The cabinet business. Like I said, you know, it's a very tough business, very competitive, a lot of fragmentation in that market. There's, you know, big, big competitors like cabinetworks and, and others that are in that space and everything down to a one man or one woman shop, you know, building one, one box at a time, right? So there's. There's still a lot of fragmentation. So, you know, the top three or so folks, you know, still only command a minority share, right? So there's. There's a lot of, lot of price pressure for that business. Very competitive and service, you know, most. I guess a lot of people probably don't realize when they think cabinets that, you know, they probably go into a Home Depot and just see a cabinet in a box. You know, you pull it off the shelf, you go to your garage, you install it. The majority of the cabinet business is made to order, right? It really is. You know, they're making this one cabinet and configuring it for a particular order, particular house, particular kitchen. So there's a lot of complexity in there and being able to take your product catalog and then make it configurable in a very unique way so that it satisfies the needs of the customers. And, you know, it wasn't just us. All cabinet manufacturers have to deal with that level of complexity. And, and really what we found very quickly, you know, kind of, you know, pivoting into the, you know, what did I find when I started there? You know, luckily, a lot of our, you know, the business that we started out was actually a carve out a divestiture from Armstrong World Industries and kind of like, hey, here's. Here's our cabinet business. This is a very common thing in private equity of a larger business, like, hey, this isn't. This isn't strategic anymore. And they find a new home for it. Oftentimes private equity firm buys it. So we got there, we landed, and it was, you know, all right, how does this business work? You know, like, what's broken? What works? You know, who are the customers? And it's a lot of like, hey, where's the bathroom even? You know, just, you know, it's trying to find Your ground. And it was, I would say, pretty quickly we realized, you know, one of the things I've always, you know, sort of been trained or more drifted towards is the customer. You know, kind of start the customer and work backwards of, you know, what do we look like from the eyes of the customer and is there any way that we can measure our performance and kind of have a good proxy for like, are we doing right by our customer? Are we meeting what we promise?

Nathan Settembrini [00:21:00]:
Right.

Michael Franklin [00:21:01]:
So that was where we started and tried to gather data. Luckily we actually had an old legacy system. It's like, God love these legacy systems. Some of them are actually pretty good. We had an old AS400 based system that had a DB2 behind it that had an incredible amount of like raw data there to work with. And you know, kind of we iterated several things and, and were able to come up with a metric that helped us track sort of our first really important KPI, which was on time complete. And it was, you know, did we, we gave a promise to a customer of when the product is going to be delivered to them and we also made a promise of it's going to be delivered complete. So I remember like in that business especially because we're, we're just a, A, you know, a player in a much larger either project or value chain. Like if you're building a home, you know, we have to be in there before other trades like the countertop and other folks can come in and complete the home.

Nathan Settembrini [00:21:58]:
Right.

Michael Franklin [00:21:58]:
So, you know, this whole timing aspect and completeness aspect ends up being extremely important, more so than price even that we found. So that was, that was really where we started. Nathan is, is trying to. All right, let's, let's measure these orders. Let's make sure every order goes out the door on time and as complete as possible. And that was really, that's where we started. We started building more and more tools around that concept to where eventually we matured and had, you know, kind of at all those major points on the shop floor, some type of scorecard or report that gives that like real time feedback to the shop floor that understand like, hey, at the top level, we're measuring on time complete for like an entire day. And then, you know, you can click on that and it broke it down into like more manageable, manageable parts. And then eventually you'd have like localized dashboards to help people manage their local area. And it all rolled up. So it worked very well. It was pivotal. And that whole concept of managing, you know, we tried to make It So it was, you know, almost no training needed. If it was red, something was bad. You clicked on the red thing and you figured out why it was read. Right. We really tried to make it easy so that people could spend their time solving rather than trying to decode some who knows, archaic report, which I'm sure you've seen all number of. When you inherit stuff, you're like, man, you guys couldn't have made this more complex if you tried. How about just putting this in plain English?

Nathan Settembrini [00:23:31]:
Yeah. So in terms of data sources, so that AS400 system is effectively acting as your ERP, is that.

Michael Franklin [00:23:38]:
Yeah. So the AS400 was the ERP again. It was your classic legacy green screen. The whole deal had a few web apps, but it was green screen. And we took that and the first iteration was, hey, can I just quickly layer over some tools just to proof a concept? And then once we're like, hey, this is working, let's start building a more traditional data warehouse and start kind of, hey, we have real time reports, but we also need to start gathering up some more data warehouse type of stuff, put it in some OLAP cubes, that kind of solution. But yeah, no, that system did really well for us and we got a lot of value out of the old. And that was one of the things coming in. A lot of some of the folks at AIP were like, hey, we heard maybe, I think they heard this from the former parent of all this old AS400. It needs to be thrown out. It's terrible. And that was one of the first things we looked at is, hey, is this thing actually, is this fine or is this workable? Right. I'm much more of a type of, hey, if it's not broke, let's not go through what ends up being usually the most traumatic thing for an organization ever, which is an ERP change. And let's just make it work. I'm pretty sure it works. And that was one of the other things we proved right away is, nah, this thing works just fine. People need to use it. Which is a common. No matter what ERP you use, it's often underutilized in many areas.

Nathan Settembrini [00:25:08]:
Yeah. So I imagine that system was tracking everything from your customers, their orders, bill, materials, inventory, like all of that stuff. Is there anything. Are there any other, like adjacent systems that tracked other things?

Michael Franklin [00:25:27]:
Yeah, there wasn't sure there was an adjacent system. Good question. So we had a CRM. Salesforce.com was the CRM CRM. So that was, that was essentially the front end where, you know, Order opportunity management, lead management, you know, kind of captured that, that initial backlog and contracts and stuff like that. And then once those contracts were won and those contracts started turning into orders, they would then flow over to the AS400. Right. Just a real simple integration between the two. And that was it. It was, it was luckily a very simple environment. I know, hey, you know, we've all been there. If you inherit something like, ah, it's just, you know, everything's everywhere. But luckily in that case, everything was in one spot. Right. And you know, the core, the system of truth was really that AS400 and then it was augmented by a CRM. Salesforce.com worked really well.

Nathan Settembrini [00:26:15]:
Nice. Yeah. I was going to ask you if the integration was a swivel chair integration. Type it over here, type it over there.

Michael Franklin [00:26:24]:
You have somebody like, oh, okay, now I got to go integrate. You know, sometimes that is the right answer, at least temporarily. But this was a more traditional event based, you know, sort of. Some of it was scheduled, some of it was event based. Yeah.

Nathan Settembrini [00:26:39]:
Yeah. That's interesting. Cool. Yeah, I. You talk about sprawl of systems. We did a discovery for this SAS company and you would not believe. So in our discoveries we start to map out like, what are all the source systems? Where does the data come from? Right. That's kind of like the second question after we understand the business and the stakeholders and what they want to know. But we start mapping out all the, where all these source systems are. And there were like dozens of point solutions that this company had kind of cobbled together. And I think that's kind of like a modern problem that a lot of companies face.

Michael Franklin [00:27:20]:
It is, you know, and it's been a problem for a while. So many of these SaaS solutions, they usually do one thing and they do it really well and they're easy to buy, easy to sell, all that kind of stuff. And very little consideration is given to how the hell are we going to make this thing fit inside the application portfolio gracefully. Right. What do we do with the data? This data now has to stuff somewhere. Yeah. So that's a problem and it's one frankly we faced later on at ACPI when we kind of stabilized that main campus and really started growing and getting some traction. One of the next things we pivoted towards was a pretty aggressive M and A strategy. Right. So again I said it's a fragmented industry. There's lots of opportunities out there to, to buy other businesses and other brands and, and kind of put them under a single portfolio. But with Those M and A acquisitions, as you would imagine, came all different Systems, all different ERPs, you know, everything was, you know, so we ended up at the very end of it when we sold, there were, I want to say, 21 manufacturing facilities. And I won. I think we had nine different ERP systems kind of along with, you know, their own ecosystems. And so that was, that was one of the, one of the initial challenges after our merger with MassCo Cabinetry. You know, we combined ACPI and MassCo Cabinetry together to form Cabinetworks. That was one of the first tasks at hand was how do we, how do we, you know, we knew and everybody had a dependency on good data to make data driven decisions. And that initial, like how do we build a data warehouse and how do we build a business intelligence platform that really can like scoop up everything as much as possible, standardize it, kind of make it all the same language and then do something meaningful with it. And some systems, like not all systems could provide the data for an on time complete report. So that effort spawned a whole bunch of other side projects of, okay, either they need to pivot over to one of the core ERPs or we need to figure out how to, how to get that data, the on time complete data because it was so critical to overall success.

Nathan Settembrini [00:29:28]:
Yeah. So how did you approach that? So when you guys would acquire another company, I guess the decision could be like, hey, let's get them onto our ERP system versus, you know, that's too traumatic. Let's just integrate the data at the BI level.

Michael Franklin [00:29:46]:
Yeah. So sort of two things. So you're right, it is very traumatic. And we were in such, we were in pretty much buy mode and the order of the day was he, hey, let's, let's accumulate these, let's, let's find some tools to help us consolidate so we can produce consolidated financials. We used a tool called OneStream. Worked great, great people behind that tool. I would use them again in a second and then sort of like one stream was our financials. And yeah, like that was every time we bought a company, one of the first things was how, how can we, you know, start putting their backlog, start putting their order into a data warehouse so that we can try to, you know, look farther and farther across to understand how the business is operating.

Nathan Settembrini [00:30:28]:
So real quick, just to kind of go back a second, when you joined ACPI, had they, had they already acquired that company? Like how, how far into the PE journey was?

Michael Franklin [00:30:40]:
Yeah, so it was about six months after the purchase, about six months after the purchase from Armstrong that AIP were kind of looking to round out the rest. You know, they felt like they had a real good handle on the operating agenda. There's like, strategic agenda as well as, like, hey, here are the, here are the players we need. Right? Like, at the end of the day, it's people, right? It's always going to be people that do this work. It's, it's, you know, you have to have the right people sitting in the right seats kind of thing. And so that's. I came in about the same time as the cfo, the COO and some other folks and we kind of all landed, which worked out great. We all landed in there about the same time. Honestly, I think my. I started on the same exact day as the CFO and CEO. So it all worked out pretty well. We're all newbies together. And then, yeah, so we kind of, AIP brought us in, said, hey, here's what we have, here's what we think, here's our thesis. And then started putting the shoulder against the wheel and started pushing.

Nathan Settembrini [00:31:38]:
Okay. So you show up and immediately you say, okay, well, where's all the data? It's inside this erp. Let's start digging around. Like, what was the, what year was that, by the way?

Michael Franklin [00:31:49]:
And that would have been like 2012, 2013. It's one of those two. Probably 2013. If I had to go back, put a date, I think it's 2013.

Nathan Settembrini [00:32:00]:
Okay. And so you show up, you start pulling the data out. Were there certain tools that you were using at the time or was it just like, hey, export a CSV for me and I'll play around in Excel and we'll try to like, get our arms around what we're looking at here.

Michael Franklin [00:32:13]:
Yeah, we really tried to. You know, my thing was I wanted to keep it as simple as possible. Right. I didn't, you know, I didn't want to add too many tools in just because, hey, that's more things to break. And it's also more things for new people to learn. So really we kept it pretty simple of we were able to use native tools or tools like SQL, ssrs. Just old school report, right? Static, static report works great. I still love it to this day. It has value in its little niche and that was really what we used. We used SSRS to connect directly to the DB2 and DB2 is a SQL compliant database. And so that was really the start of it and kind of views and store procedures and we built a few offline things that required a little More processing. We didn't want to interfere with the transactional system, which is common. Common problem we have to all of us have to overcome. And that was really where we, where we kept it is let's try to keep it lean and mean something we can quickly iterate over because we were again, you know, like anything we weren't. There wasn't just a one person had a vision of they could draw it on a whiteboard of, oh my, you. You guys go and build this thing and it's perfect. It was more of a, hey, we think we need to manage like customer deliveries. What. How do we do, you know, and we had to really decode that of all right, well on time complete is what that's called. And then, all right, what data is available that we could even use to measure it? How could we measure the details below it of, you know, any order gets exploded in the bill of materials down into a whole bunch of messy work orders and subcomponents and all these things. And so that was, that was one of the. Honestly, that was probably the biggest learning curve is understanding that schema of like, hey, there's a whole bunch of data inside this AS400. What is it? Where is it? What does it mean? You know, talking to the vendor that made that was behind the ERP system to try to understand we have this concept, we want to build these dashboards. Where can I find the data? What, what is its format? Is it in a format that I can use for reporting? Or do we have to like manipulate it and do some ETLs and such with it? So that was, that was really the first, probably six months, nine months is kind of drafting those first iterations and then, you know, kind of standing up the top level and then the next level down. And then, you know, we kept. We kind of that top level one and then kind of branched out of okay, here are the major areas. And we, we just tried to drive it down as. As much as we could. So that the idea was to put a tool, an information tool into the hands of a shop floor operator or a foreman that they would know exactly like what was left to complete. Was there anything that fell out? You know, like sometimes material in a manufacturing process gets damaged, has to be remade, those kind of things. And so that was the idea of how can we make sure everybody has a tool they need to be successful. And then kind of one of the final things which proved to be one of the most valuable is building a set of analytics that took a lot of that same data and Then gave and put it in a format so that the shop floor understood how to that day's production equaled what they were likely to earn on their performance bonuses. Right. Like I've, I've worked in a lot of manufacturing environments where if a performance bonus system is structured correctly and it's, it's this win win type of situation where you really allow shop, you know, the shop workers and leaders to kind of control their destiny for the day, you know, and really, you know, make sure that of the things they can control, they are controlling it. And you have compensation that aligns with that. It can be extremely impactful, very powerful. Yeah, and we did that and that was one of the sort of the last, like, all right, that was the last one of those reports that sort of slotted in and said, all right, this thing's great, go. And we, we tried to use that same format on all the companies that we bought. That was one of the first things we tried to do is like, all right, can I shoehorn, can I take this in thing and stuff it with data from this new system that we just inherited? And for most of the businesses we bought, that was possible, at least to some degree. Was it perfect? No. But it was close enough to where it made an impact. It made a difference.

Nathan Settembrini [00:36:42]:
I love what you said, kind of implied and what you said is a lot of times people will focus on the technology solution and forget about the people and the process. Right. And so you identified this metric really matters to our customer. Therefore it matters to our business. And in order to grow our business, we need to satisfy our customer. And in order to do that, we need to operationalize something, change our process and incent our people to, to do the things that then make our customer happy.

Michael Franklin [00:37:21]:
That's right.

Nathan Settembrini [00:37:22]:
Yeah. That's awesome.

Michael Franklin [00:37:23]:
And then, you know, everybody, everybody's, you know, going, everybody has clarity of. I think, you know, clarity is a, is a underutilized word in business. Of. There's too often businesses lack clarity in what is important, what are we working on, you know, what actually moves the needle and when you can get those things right and then it's easy to communicate to the business down all the way to a shop floor worker. Everybody, like, needs to understand why is, why is this thing important and why does everybody care if it's not where it's supposed to be? Getting that right is so important for a business.

Nathan Settembrini [00:37:59]:
Yeah, yeah. Are there any. Okay, so as you guys continue to scale out, you would kind of take this framework that you've developed that you Know, works well apply it to the new company, roll everything up. Like what take us through the journey kind of from that point to the acquisition or the eventual exit.

Michael Franklin [00:38:19]:
Yeah. So really at one point we had grown ACPI. You know, we had bolted a lot of acquisitions underneath ACPI. I think at that point we're right around 8,900 million for just the ACPI side. And we kind of had an opportunity that the PE firm wanted us to take a look at which was sort of what would be classically called a merger of equals of hey, ACPI has reached this milestone. We could continue to buy smaller firms or we could just partner with another firm that a large company, Masco would wanted to part with and form what ended up being the largest private cabinet company in North America. And that's essentially what we ended up doing. That would have been like the 2020. It was, it was around Covid. It was a kind of weird time. But that was, that was when that merger happened was in the 2020 era timescale and that was, you know, AACPI sort of cease being Masco. Cabinetry was no longer and the new one was Cabinet Works Group. And that was, that was kind of what we carried forward and eventually took through a process and sold to a PE firm called Platinum Equity and they operate it still today.

Nathan Settembrini [00:39:32]:
Amazing. At some point did you build out a business intelligence function underneath you?

Michael Franklin [00:39:41]:
Yeah, yeah. So that was quickly. It was one of those like all right guys, I'm happy to write code, but I got other things to do here. I do enjoy writing code. But yeah, we had to very quickly pivot to hiring some folks to help data engineers. We found a third party team to help us marquee marquee data that, that was instrumental in, in helping us sort of mature that practice. Right. Of I'm not, you know, hey, I, I, I got us to a certain point, but I'll be the first to admit I'm, I'm not an expert, you know, building out data warehouses and business, you know. And so that that group was able to really take it to the next level, mature the practice. We actually brought in some advanced analytic tools from Tableau that, that you know, hit the mark really well for a lot of more like ad hoc analysis, you know, to put those tools kind of in the hands of true analysts that you know, you know how it is. So we, we maintain the static like hey, you know, dashboards and reports, but also started adding capabilities on the more visualization as well as as sort of traditional analytical tools for like a system Or a business analyst would use.

Nathan Settembrini [00:40:51]:
Yeah, yeah. So the general process was hire an external firm to help us build the kind of the core data warehouse and then start bringing in team members and tools to kind of self service analytics with tableau on top. Did you start with a leader like your first hire in the BI space? Was it an analyst? And then you kind of built up from there. Do you start with a leader and then.

Michael Franklin [00:41:17]:
Yeah, great question. So, you know, looking back, I, I probably should have. You know, if I had known, we would have ended up, you know, this large 1.8 billion dollar. Yeah. I would have picked the BI leader and just like let, let him or her build a practice. Yeah. But you know, as it, as it sort of grew organically, it was, hey, let's, let's put a data engineer in here. So that could like take all the, take all the actual development, the back end development off me. And then, you know, he needed some help. We plugged in a third party, like, all right, let's graduate this to you know, sort of the classical multi tiered architecture. And then when we eventually merged with Masco, they actually came to the table with their own, their own business intelligence apparatus as well. Like, you know, fairly mature, based in Oracle, Obie and it worked well and they were used to using it. And so we kind of, for a period of time. All right, let's just, we'll maintain these two. And we kind of layered over a consolidation piece on top that kind of sucked data out of those two different data warehouses. And since then, since I've departed, it's been three years, they've consolidated. That was the long term goal is, hey, we have good data over here, we have good data over here. We need to consolidate and collapse this into just one maintainable data warehouse and a set of data models that are the source of truth for the organization. And that's, that's where they've ended up today.

Nathan Settembrini [00:42:43]:
Wow.

Michael Franklin [00:42:44]:
Yeah.

Nathan Settembrini [00:42:45]:
Yeah. So we've talked a good bit about how, you know, the company matured over time, how the, how the solution to find basically to uncover data secrets. You know, this is the data secrets podcast after all. Right. And so you've got the process in place, you've got the people, you've got the technology now. Like, are there any stories that you remember where there was some, something that was like a problem in the business or it's like, golly, we can't really figure this out. But then the data really like finding the right KPI or finding the right insight from the data led you to make a big leap on the business side.

Michael Franklin [00:43:31]:
You know, that's a great question. There are some certainly like all things in life, you know, you always get surprised. And I think one of the, one of the interesting surprises that we ran across is while we were over here focusing on OTC on time complete and building all those operational things on the sales side, on the customer side, we had this assumption that, you know, hey, the biggest customers are the best customers and you know, those are the guys carrying the guys and gals carrying the sort of load quote unquote for the company. And, and you know, someone, I think it was one of the, the sales leaders. Like, you know, I'm not sure this, this is, this assumption's 100%. Actually. It came from one of the AIP partners, come to think of it, and they're like, you know, I'm not sure your guys assumption is 100% correct. You know, is there any way we can start looking at, you know, building some kind of contribution margin, basically what ended up being a customer segmentation type of analysis or data model, you know, and, and like, like you would expect in any type of. Once you build that, you start looking at it. Yeah, there were definitely accounts that were large, they had a lot of volume, but they were far lower margin. Once you actually built up a contribution margin, they're far lower margin than the business realized, right. To some degree, some of them were negative margin. And so that was one of those surprises of, okay, I'll probably use that same make sure that question gets asked earlier, you know, in every subsequent engagement and endeavor. But yeah, that was, that was surprising how we, you know, we were dead set, me included, of like, oh, that customer is. Those guys are great. And when it really came to stacking it all up and putting, you know, putting everything in there like they're nice people, don't get me wrong, but they don't make any money for the business, you know, so it's like, hey, we had to make some, some tough choices there because at the time it was, we were very capacity constrained as well. And so it was like, hey, we need to make sure we're doing the most we can to sort of maximize profit. You know, it's a cyclical business. Like a lot of manufacturers, everybody in building products is this way of, you know, you need to, you need, when you can do well, you need to do well because it'll go the other direction on you as well. So that was that. That customer segmentation model is probably the most interesting bit of work. And it ended up sort of creating its own set of analytics and tools. And eventually we got it to a point where every, as soon as every order was imported and brought in, management knew within minutes of what the expected contribution margin was going to be for that order. And we had some sort of logic, like tree logic to sort of, hey, you're not going to like, hey, you've already accepted the order, right? You've already taken. But it's more like, hey, we need to have some people take a look at this account and, and, and revisit it. So, so that was, that was probably the, the biggest secret that there was that unlocked a whole bunch of, of, of margin that was just leaking through the bucket. You know, the old, like, hey, where, where's our bucket leaking? That was, that was a major one, I would say, you know, probably that's the most, you know, the other Can Am Steel had a similar situation to where it was more the inverse of management felt that a situation existed and we used data to help validate that, which was, hey, we felt that the more complex jobs were really good at dealing with those complex jobs and we felt that they were more profitable because we were able to command a higher price for them. And that was another case where with some analysis that has proven to be so. And it's actually even carried, like that's part of the core strategy going forward for that, that firm. Yeah.

Nathan Settembrini [00:47:23]:
Nice. Nice. What about. So if we, if we look at the dark side and go dark here for a second with the data data, horror story. You know what, Was there ever a scary moment or, you know, a decision that was made on some, some bad data?

Michael Franklin [00:47:41]:
Right? Like, you know, mistakes are always the best teachers, aren't they? And, and they're humbling as well. They have that benefit. You know, at Capital Works, we definitely had a mistake like that. We had built this beautiful quoting tool, you know, salesforce.com based, all this great, looked great, worked great, very fast, very reliable, and you know, put it into production. Worked for weeks, actually. I think it worked for a couple months. And then we realized like, hey, this like the logic's all wrong, you know, and it was one of those, like, hey, we thought it was wrong. Some edge cases came back to really bite us in, in the, you know, and, and it, you know, so that was one where we ended up underpricing a whole bunch of, whole bunch of jobs, you know, and there weren't any alarms, there weren't any flags, there weren't any business rules that like, gave us visibility to that it was just sort of like, hey, that thing works great. Let me know how it works out for you. And. And it wasn't until some of those things started coming into our backlog and. And it was sort of under, you know, those tools that analyze the backlog is where it started coming out. It was a gut punch. They had to, we had to stop quoting. We had to scramble to fix the logic. And we were very customer focused and we weren't going to go back to those customers and say, oh, sorry to pull the quote out from underneath them. It was, no, this is something we just owned. We took the hit and tried to learn from it. And it actually ended up being one of the catalysts that we created a system that we, we called the, the Business Process Management Business Process Monitor system that really, it was just a rules engine. So it was, hey, this thing was constantly taking rules that we clearly identified though essentially it just overlaid guardrails over a lot of the process. And we expanded it to kind of all over the business after a pretty short amount of time to give us that confidence of like, hey, it's great, we're going to go full forward and move as quickly as we can. You know, under that, you know, very much that fail fast and fix mentality. You know, a lot of PE firms share that same mentality as like venture capital is, but we're not going to waste this opportunity to not, not learn from this mistake. And that BPM system ended up being awesome. Like it was. It was. I'm not sure we would have without that need. I'm not sure we would have come up with that. But it ended up being extremely valuable.

Nathan Settembrini [00:50:04]:
Nice. Did. Did the quoting tool misquote because of some technical issue or lack of business understanding or like, what was the root cause of why that failed?

Michael Franklin [00:50:16]:
I think it really, if you come down to it, it ended up being. We didn't fully understand all the use cases. Like, you know, if you were to say, hey, we thought it was these 10 user stories. Those user stories actually had like someone, you know, it was a lack of understanding of the actual process. And when it got interpreted by. We had a third party do a lot of the development for salesforce.com and it was, it was, you know, kind of looking back like, well, they delivered the tool exactly as we asked them to deliver it. The problem was it didn't actually account for all the situations correctly.

Nathan Settembrini [00:50:49]:
Yeah.

Michael Franklin [00:50:49]:
So it was, it was very much a. If we had a, you know, if we had to do it over again, it was A lack of diligence. It was a lack of, you know, sort of putting in the time on really understanding all those use cases. In reality, Instead of the 10 user stories, there were probably about 20. You know, I think we. We got about. We. It's like the 80. 20, right? Yeah, we nailed those 80. And there was some there. The trouble was in that 20 that we just really didn't even know truly existed because we were moving fast, you know, and that's. That's. That's what happens when you move fast. You're gonna make mistakes.

Nathan Settembrini [00:51:22]:
Yeah.

Michael Franklin [00:51:22]:
And this is one where. Yeah, that one. That one sucked. That was a mistake. We had to eat for a little bit of time.

Nathan Settembrini [00:51:28]:
Yep.

Michael Franklin [00:51:29]:
Yeah.

Nathan Settembrini [00:51:30]:
All right, so let's. Let's pivot to the kind of what you're doing now with Cold Iron Capital and thinking about. So a lot of times when people think about making an investment in technology, or BI specifically, right. You're gonna. It's kind of like plumbing, right? It's like you got to build the thing, and then you start to use the thing in order to see the value from the thing. And so sometimes that ROI is. It's hard to promise on the front end because it's. You don't know. That's why we called this podcast the Data Secrets podcast is you don't. You don't really know what the data is going to be able to tell you until you do it, until you invest in it. So how do you think about investing in data solutions and driving ROI from those?

Michael Franklin [00:52:20]:
No, awesome. You know, great question. Because as you know, any of these projects can get extremely expensive, like, really quickly. Right. And this is before you even buy all the tech. So I think I like to think of it no different than any other capital project. You know, the tool, the endeavor needs to solve a specific business problem. It needs to unlock throughput, improve margin, reduce risk. You know, it needs to solve something that's already been identified as a problem. Right? Like, you know, like the classic saying is, you know, a problem that's. That's. Well, it. Well defined is halfway solved. And I think, you know, not enough. And certainly I'm guilty of this in my entire career, not, not spending enough time really defining that problem, really understanding of what. Like, let's be firm on what are we trying to solve. And then basing and sort of jumping off from there of, hey, if. If we. Well, if we understand a process and, you know, whether you use Lean or theory of constraints or something, some mental framework to understand, like, hey, here is the constraint we're trying to alleviate. Or here's the. If we only knew this information, we would more consistently make better decisions. That is where those projects need to start. And I'm always much more excited about approving and underwriting those kind of projects or firms that are clearly well defined. The problem that they're actually going after and trying to solve. Too often it's just, hey, teams know they need to put in business intelligence. So go do that.

Nathan Settembrini [00:53:53]:
Build me a dashboard.

Michael Franklin [00:53:55]:
They'll build me dashboards. And it's like, well, you know how it is. Well, I can put a lot of stuff on a dashboard. It might not be useful, might not tell you anything, it might not change anybody's behavior. And so if, if a KPI or a tool doesn't fundamentally change behavior of the person using it, it's probably noise, right? Like, that's the thing it's supposed to do is it's supposed to trigger some kind of action and a decision. Right? Or at least, at the very least, hey, I know this thing is still nominal. It's still in flight where I expect it to be like, that's a, that's an okay use for a tool too. But yeah, if we're not really attacking a real problem, man, those, those just almost always end up being a disappointment on a whole bunch of levels.

Nathan Settembrini [00:54:43]:
Yeah, yeah, I feel you. Is there anything else you want to talk about on your kind of your investment thesis or like what you're planning to do over the next five, ten years?

Michael Franklin [00:54:53]:
Yeah, no, man, this next ten years I think I'm actually really pumped. So obviously a lot of my energy going forward is going to be cold Iron Capital and some other sort of similar endeavors along the same line. But as I look for early stage or product based firms, I'm really looking for people that are doing again, solving some unique problem. They're trying to do it with intelligence systems, you know, whatever that. There's a whole bunch of AI. People say AI, but there's a whole mess of, you know, there's, there's a whole bunch of stuff there, you know, data in those spaces of. I've always really liked both. I, I like data. Intelligence Systems is another one. Cyber, you know, security itself, I think I, I think, you know, kind of pivoting forward of. Yeah, I have a lot of interest in this venture capital and I think there's some interesting things going on there. But I'm also really interested. No one talks like we're all talking and rushing as quickly as we can to integrate AI. But there's a ton of security implications with that. These projects, as they get implemented, they're all new code bases. A lot of them are wrappers and things that are done very quickly. And so there's really not a week that goes by that I don't see some poor vibe coder that launches a tool and like, oh man, I've made success, I'm getting traction and the poor person, you know, ends up exposing their S3 bucket because they didn't know any better, you know, and it's like, hey, yeah, good for you guys for delivering a product and you did something that people want. But because either either you or the tool lacked that overall context and know, like, hey, you know, you're not quite there yet. The tool works, but you need to, you need to really operationalize it. So there's, there's a bunch of interesting companies, obviously the players like AWS and Microsoft, Google clearly making a lot of moves there, you know, tools. There's also some startups like LangChain that are trying to build some specific tools to try to address, like guardrails, like how do we, once an AI system is deployed, how do we ensure it sort of stays inside of some given parameter set and doesn't drift over time, which is a common problem. So there's a bunch of stuff I think that's really interesting there that just personally I'll be spending more time with in this data space though. One thing, I don't know if you guys have seen it as much, but I've seen it with other firms that are dealing with a lot of analytics is rag, the retrieval augmented generation. Hey, there's this new sort of database that's now being sort of brought into a lot of data stacks to facilitate these gen AI tools and workflows. And these RAG systems are vector database systems that, you know, SQL servers trying to add one but like Pinecone and others are out there that have, like Supabase has one now as well that. All right, you know, companies are rushing to implement RAG because that's the, right now the quickest way to really cram a bunch of proprietary information into a database that then an LLM like an OpenAI can go and fetch information out of so they can provide better answers, you know, specific to a business. But it, you know, so yeah, that's cool, that's great. Those are fun to build, those are really interesting. But it just. There's so many, there's so many things to learn. There's so many, you know, what are the best Design patterns, What are the right ways to do that going forward? So, like, that's, that's the frontier. That's still. That's still to be determined. And so I'm really interested in how that stuff evolves and where that goes.

Nathan Settembrini [00:58:32]:
Yeah, it's interesting. You know, there's this concept of AI readiness, right. And so, yeah, effectively AI readiness is traditional bi. It's like get all your data into one place, right?

Michael Franklin [00:58:46]:
Yeah, it's one place. Some kind of schema that I can define it.

Nathan Settembrini [00:58:50]:
Right?

Michael Franklin [00:58:51]:
Yeah, like, you're absolutely right. So I think, I think firms like, like yourself and, you know, there's. There's a huge opportunity there of so many firms still. You know, like those firms that put in that hard work of laying that foundation over the last 10 years or five years, whatever it was. Oh, man. They're going to have like, such an advantage going forward because they're already starting from a foundation that isn't super expensive or super difficult to start from. And I think the smaller firms actually have an advantage there in a lot of ways. Right. Like a huge mega firm that just is, you know, 10, $15 billion super complex. Data is everywhere. You know, I'm sure you've seen those. It's like, wow, I don't know how you're going to get all that stuffed into any AI, you know, but the smaller firms, they're smaller competitors. If they're smart about it, I think they can just move a lot faster. Right. And like, speed kills. So I think there's some small people, they're going to see that and realize, like, hey, that we have the data and we're able to have the right set of people or partners available to do something with it, and they're going to be able to move forward rapidly. Much, much more rapidly than their competitors anyway.

Nathan Settembrini [01:00:05]:
Yeah, yeah, yeah. I kind of think about, like, if, if you were at ACPI, like if you were just joining ACPI now, right. Or a company like that. Right. And so they're, They've got their source systems. There's no BI layer on top. I think the speed at which you could stand up a data warehouse and start providing insight is just so much faster today than it was.

Michael Franklin [01:00:37]:
It's way faster. And the key piece there that there probably isn't. Nobody really wants to talk about this because it's not as sexy either is just bolting on some AI tool to an existing process sometimes will probably yield a good outcome. But more often than not, you know, what is it? Like Bill Gates said, you know, Automating an inefficient process is just going to, you know, mag. It's just going to amplify the inefficiency. And there's, I think so many of us, you know, that have dealt with these systems over the years. We think, oh, well, you know, management just put AI on it. You know, why AI? And it'll solve just no. You know, no is pretty much the answer there. And it's. We'll need to do complete redesigns in some cases of entire processes to really make use of it. Otherwise it won't be adopted. It'll just be some bastard sort of stepchild off the side that costs a lot of money and it doesn't really get any usage, doesn't change anything. But those smaller firms are definitely an advantage, Nathan, if they can. I think it probably comes down to willingness to take some risk and willingness to hire some people that know what to do or at least hire some firms that they can, you know, pick some. Pick some existing. Again, it comes back to like, you have to be solving a real problem. Pick some existing bottleneck. You know, whether it's, hey, we. We don't have enough hours in the day to do all the quotes that people ask us to do. Okay, start there. You know. No, no business ever failed because they had too much sales. Too many sales. Right. Like, typically. So, you know, if some firms can pick the right projects and they're lucky enough or they're diligent enough to find the right partners to help them implement some smart solutions, you know, intelligent solutions. Yeah, they're gonna, they're gonna have, at least for a period of time, they're gonna have a pretty significant advantage.

Nathan Settembrini [01:02:27]:
Yeah. Yeah. One thing I've kind of observed over the past couple years is firms kind of pausing on investing in data infrastructure because they're kind of assuming that at some point AI will just solve. Solve it all for me. And so I just. Just kick the can, you know. And what would you say to a firm that's in that situation or thinking that way?

Michael Franklin [01:02:50]:
You know, it still comes down to AI is just one of the tools. That's it. It's just one of the tools that'll be inside your application portfolio. Are you going to walk away from your CRM? No. Like, the CRM provides the structure and the scaffolding and the process workflows to run that part of the business. And that doesn't change what companies need to spend their time right now doing is, hey, analyzing what is our strategic agenda versus where is our current State of processes. Right. Like if we have a real aggressive growth agenda, is our sales side really, do they really have the tools and are they empowered to meet or have a chance to meet those sales objectives, those growth objectives and start doing that work? Of all right, here's A, B and C processes that need work. We've done the work to map out the flows. We understand like we have a bottleneck here and here. And then it's, you know, once you have that point, you're in a great position to go out, find partners, find people that can help you. Yeah, maybe, maybe an AI tool exists to solve that. But you know what, sometimes you just need to redesign the process. Maybe your process just sucks, you know, and the highest leverage thing you can do is just redesign the process because that's fast and cheap. So I think like any of the, I love the tech, I love AI. I've been enamored with artificial intelligence since I was a kid reading sci fi. Right.

Nathan Settembrini [01:04:17]:
Like yeah.

Michael Franklin [01:04:17]:
And so I couldn't be more excited about this period of time. But the reality really is for a lot of companies, yeah, there's some tools that'll, that'll help you, help you win the market, but it doesn't take take place of like the traditional business improvement. You know, there's one of the processes what needs to be fixed. Do you have the right players? You know, there's a lot of upskilling like that. That is a thing that I think people really underestimate is, is the level of upskilling that you're, that everybody really will, will need over the next five, you know, four to five years of, you know, things like, although I'm not a huge fan of Copilot but like the idea is there of, you know, there should be all the native tools they're now building in some type. You know, whether it's Copilot with Microsoft's tools, Oracle has it, SAP has it, you know, netsuite, they're all starting to, you know, Salesforce has it with Einstein and some other things and their agent system. Yeah, it'll like that. That's usually where firms will find the easiest. Like the least amount of friction is to try to leverage the built in tools. Right. Like Gartner said the same thing last fall. I'm not coming up with anything new here. It's more. Yeah. If you have an ERP or a CRM that has the embedded tool set, learn it, use it, find ways to incorporate it into your workflows. That's usually your best bang for the buck. And in a lot of cases you're already paying for with your licenses. And then, you know, at that point, sort of maturity, the next level down for maturity are those point, those, those points, those use points. Like, hey, I have a use case, a specific use case that I need to. Either there's an off the shelf, which sometimes there are, but more likely it'd be I need to assemble some collection of workflow automation and LLM models.

Nathan Settembrini [01:06:03]:
Yeah. Yeah. Nice. Well, Michael, as we look to land the plane here, where if people want to connect with you, learn more about your firm, like, what's the best way for folks to connect?

Michael Franklin [01:06:17]:
And yeah, I'd love to hear from people. You know, I have, I love connecting on LinkedIn. LinkedIn is probably the easiest place to find me. I would say if you just search Michael Franklin and Cold Iron Capital, I'm the only one that pops up. So just hit me up there. I accept almost every single request. Save a couple obvious spam bots, but you know that that's a good place to hit me up. And then you can also email me. So just mike@coldironcapital.com hit me up there and Happy to start a conversation with anybody. Sure.

Nathan Settembrini [01:06:47]:
Okay. Yeah. Who, who would be kind of the person you'd love to, to hear from? Is it a CEO of a company that's of a certain size or. Yeah.

Michael Franklin [01:06:56]:
Yeah, good question. So I would say, first and foremost, I'm happy to help anybody. You know, I'm a firm believer in reaching your hand down and helping other people climb up the ladder. So if you're ambitious, you want to, you want to eventually be in a CIO chair or a cto, something like that. Hit me up. Happy to. Happy to use whatever resources I have to help the other is, yeah. CEOs, board members, you know, PE partners that are like, hey, what, what, what do we do? What do we do with AI digital transformation? Any questions like that? Happy to have a conversation again. There's, there's a lot of, there's a lot of pain and suffering that I've gone through, and there's no reason other people should have to go through the same. So. Happy to at the very least help people avoid those mistakes that I've made. And, you know, people just want to reach out and have a conversation about AI. Happy to do that, too.

Nathan Settembrini [01:07:49]:
Awesome. Well, thank you so much for your time, Michael, and happy to be here.

Michael Franklin [01:07:53]:
Appreciate it.

Nathan Settembrini [01:07:54]:
Awesome. Well, that's the pod.

Michael Franklin [01:07:56]:
All right!