Digit is a next-gen, affordable ERP built for growing manufacturers and distributors who need real-time control over inventory, production, and orders—without enterprise complexity or pricing. It replaces spreadsheets and disconnected tools with a single source of truth that’s intuitive, fast to implement, and powerful enough to scale as your operations grow.
In this episode, we speak with Dan Koukol, Co-Founder and CEO of Digit Software, about the challenges and innovations in the physical economy, particularly in manufacturing and supply chain management. We discuss the limitations of legacy ERP systems and the need for a “system of progress” that integrates modern technology like AI and robotics to improve efficiency and decision-making. Dan emphasizes the importance of education and alignment within organizations to unlock potential and drive growth. The conversation also touches on the future of operational software and the role of AI as a coach to guide businesses in optimizing their operations.
Watch the podcast here:
Listen to audio only here:
Learn more about Digit.
Interested in appearing on the SourceForge Podcast? Contact us here.
Show Notes
Takeaways
- The physical economy is often overlooked in the AI-driven world.
- Legacy systems create significant gaps between expectations and reality.
- A ‘System of Progress’ focuses on proactive guidance for businesses.
- Education is crucial for companies to understand their operational metrics.
- Speed of implementation can significantly reduce business risk.
- Spreadsheets are still widely used due to their flexibility.
- AI can serve as a coach to help companies improve operations.
- Alignment within teams is essential for organizational success.
- Data integration is key to unlocking operational efficiency.
- The future of operational software will be shaped by data and AI.
Chapters
00:00 – The Physical Economy and Its Challenges
01:13 – Legacy Systems and Their Limitations
04:34 – The Missed Opportunities in Modernization
10:47 – Introducing the System of Progress
15:24 – The Object First Model and Its Importance
19:56 – Clarity and Financial Health in Operations
23:12 – Operational Discipline and Speed in Implementation
26:06 – Unbundling Monolithic ERP Systems
28:44 – The Role of AI in Business Operations
33:12 – Transforming Tribal Knowledge into Organizational Alignment
36:35 – The Spreadsheet Dilemma in Mid-Sized Companies
39:31 – Future Structural Advantages in Operational Software
Transcript
Beau Hamilton (00:01.109)
Hello everyone and welcome to the SourceForge Podcast. I am your host, Beau Hamilton. All right, now here is the setup. Here’s the scene. We all spend all day staring at screens, but we still live in a very physical world. Everything in front of you and around you, your chair, maybe your desk, your coffee in my case, this podcast microphone, someone had to build that and they had to ship it. And that is what’s referred to as the physical economy, which you don’t hear people talk as much about in this AI world these days. All the attention kind of goes to the latest AI models.
Even though the people actually running the factories and the warehouses that make the physical things we have in front of us, they’re stuck in what’s often called and referred to as spreadsheet hell or outdated software that hasn’t changed in for many cases, you know, decades. So my guest today is trying to fix that. I’m sitting down with Dan Koukol, CEO of Digit. Dan isn’t just a tech founder. He’s a former manufacturing CEO who has actually walked the shop floor. He’s walked the walk, talked the talk. He knows the struggle and the legacy of ERPs firsthand.
And we’re going to talk about why the old system of record is broken and what he calls the system of progress, which is a new way to help the physical economy actually move forward. So Dan, welcome to the show. I’m glad you can join us. have a lot to talk about.
Dan Koukol (01:13.528)
Yeah, thank you for having me, Bill. Excited to dive in. I always like talking about next generation ERP.
Beau Hamilton (01:18.323)
Yeah, I’m excited to get into it now. Dan, I understand that over the course of your career, you’ve had countless discussions with manufacturers and distributors before you sort of stepped into the software space and you actually taught yourself to code, right? That ultimately helped you move into this space and helped you solve some of the problems you were noticing with outdated operational software. Maybe you can start by talking about that disconnect you were seeing. What’s the biggest gap you saw between what software said it would do and what people actually felt on the shop floor.
Dan Koukol (01:52.3)
Yeah, think there are a lot of issues with legacy systems and software. And like you said, I lived it firsthand. So my career started as a consultant, worked for nine years through dozens of manufacturing companies, distribution companies, and actually ran a manufacturing company myself. so during that time, I’d used any system you can think of, I probably used it toward a part, said what’s good, what’s bad, trying to get those systems to help me manage, either isolated parts in these manufacturing companies or the entire supply chain end to end. And a few things that were problematic early on were number one, the original promise of these systems sounded really good. Run your entire supply chain on this, but the actual application of that and trying to get the products themselves, the software products to work end to end, never really came to fruition. There was a lot of customization that was needed. There was a lot of like, hey, let’s try to do this workflow from sales order entry through fulfillment, let’s say.
But there was something unique about the business that just didn’t allow the off the shelf to work. So there’s a lot of pain and just trying to get it to work. The other thing, and this is a big part of the thesis, the founding thesis behind Digit is even if you could get all of that stuff to work, it wasn’t enough to get everything on one platform. If you did that, you still have the operators, the users, the management needing to go in, know what questions to ask and mind data from the system to try to figure out what they need to do to improve their business. And for me, that just never really sat well.
If you have all the data, if you have all that system, the system should proactively push that information to the different stakeholders in the organization, guiding them into what they need to do better based on the data that system has, the knowledge it has about the industry, the size of company, whatever it may be, to teach management and operators what they can actually do to improve. those two big things, think, were the biggest gap I saw was, you know, lack of ability to actually install the system, it working, and then the lack of, you know, proactive guidance that those systems can provide.
Beau Hamilton (03:46.035)
Yeah, that is something I hear from a lot of people like yourself in the industry where there’s all these dashboards, these checklists of features and whatnot, but they don’t actually translate to meaningful results on the factory floor with the people that they’re intended to be designed for. Now, when you look at the last, let’s say, few decades of technology, there’s been a big increase in all sorts of things like data collection. You have this kind of general move to a centralized control interface, all in one approach with financial reporting tied into that. But it seems to have largely stayed pretty consistent from there. Where do you think the industry missed its chance to truly modernize operations and supply chains? Was there maybe a particular moment in time that kind of stands out to you?
Dan Koukol (04:34.168)
Yeah, I think the real transformative shift happened in the early 2000s when cloud came to be, right? So you had mainframes, you had a lot of on-premise software that existed prior to. And while cloud was mainly a difference in medium and delivery model for how the software was distributed, they had a chance to re-architect a lot of these products to take advantage of what you can do with cloud, right? So you can push updates very quickly. You can update the UI, the UX. You can make things more configurable. And so there was, in our view, this really kind of golden era of opportunity to solve a lot of the pain points that were causing the legacy systems to not be adopted, which are the user interface is overwhelming. There’s a hundred buttons on the different screens. It’s too clunky. can’t get it to meet my workflow, where you have a lot of other kind of
Point Solutions or consumer grade apps that took advantage of all of this really started investing in the user experience. That didn’t really happen in B2B software specifically, not in the manufacturing and supply chain area. This was kind of unsexy. The companies who really came and dominated the cloud didn’t really, in our opinion, solve product. They didn’t create a beautiful product that had a rational brand loyalty or product loyalty from customers. It was still painful. It’s still very clunky.
So I think that’s the big one. And then the second piece that I think was really missed is a lot of these companies were told to find a niche in focus. And so what you have now is this sprawl of individual point solutions that manage different pieces of the business. And you had fewer teams really take on the challenge of building a true large scope product where you could do things in one place. And so that SAS sprawl has really existed for the last 20 years.
And that’s where you have companies using a multitude of different systems to manage different areas of their supply chain. So I think, you know, not nailing product and also, you know, only really having point solutions to choose from have been the biggest things that have held software back in the space.
Beau Hamilton (06:34.613)
That makes sense. It’s a good explanation. know a lot of the big ERP systems, really like SAP, NetSuite, and others, they really can establish themselves in that time frame. And they’re fairly ubiquitous now. like you’re saying, they’re part of the problem you’re getting at, where maybe, I don’t know, they stopped kind of innovating. maybe they stopped like, there’s just that disconnect. They stopped maybe listening to and
tracking that kind of inefficiency or that they notice that area where there’s not really, they’re not seeing the results translate from the software implementations and features that they’re implementing. Like, I’m curious, why is it so hard for those systems to deliver just the basics, like simple, fast, usable experiences for companies that are growing?
Dan Koukol (07:24.514)
Yeah, think the real answer is there’s probably eight or nine things. It’s very layered, but I think at a high level, what you have is a lot of the origins of these legacy systems start with code bases that are just kind of built on top of themselves over time. And you don’t have a lot of these more modern, you know, when you think of a product like Notion or some of these more modern apps that have workflows built in, that have configurability from non-technical users, custom fields, things like this.
those weren’t really done at a high level early on in these product life cycles. And you had a lot of customizations being built in where if you’d go to this food and bed manufacturer, this company or this company, you would kind of custom develop solutions to meet the needs of those companies. And that ends up getting your product to be very kind of unruly and messy. There’s also some ways that these companies have gone to market where they have implementation teams and third party
ecosystems that monetize the customization of these platforms. And so I think when you look at all of that together, they find that they are installing their product in a of companies, they lose the ability to now upgrade or improve their product because they’re locked in with all of the different custom implementations and their architecture now has reputation. hear a lot of people
you know, it’s rigid, it doesn’t work, we need to pay $15,000 to add a column in this report or we need to do something like that. That’s a result of, think, product strategy and roadmap management from the very beginning, where early on, if you can build different objects like sales orders, purchase orders, all the things that make an ERP run, and you can make them industry agnostic, work for SMB, mid-market and enterprise, and be really intentional and disciplined early on.
and you can build those primitives, you can build a custom workflow builder, can build a custom form builder versus building like a food and beverage batch tracking specific application. Those are the things that allow your product to be a lot more flexible to be applied across use cases where you see now a shift from customization, hard coded customization into now non-technical users configuring the system, not customizing, but configuring it with drag and drop.
Dan Koukol (09:42.656)
things like that that you see more modern systems.
Beau Hamilton (09:45.375)
Right. Yeah, I feel like we can have a whole kind of episode just kind of dissecting where they kind of went wrong. But I think a lot of the customization aspects and trying to move everyone towards this one approach as opposed to the custom kind of tailored approach, think is definitely that makes sense why there’s some issues surrounding that. But also just some of the things that made them powerful in the first place. You think of just the scale and configurability.
financial control, those things, like you were saying, they make them just slow and a little bit rigid for operators. And I want to get to the system of progress term that you referred to, digit, kind of implementing it and really following, I guess. And why don’t you refer to the company? I know you don’t refer to the company as an ERP in the first place. Talk about what that system of progress means and why it’s so important to move beyond the idea of
of a system of record to help drive business growth.
Dan Koukol (10:47.502)
That’s a great question. Very passionate about it. So the reason we don’t, you know, I’ll use ERP here and there, but ultimately at our core, we don’t associate with ERP. think number one, ERP is useful in that a lot of people can kind of understand what that is. So when you say ERP, people have some kind of basis. At the same time, ERP means a lot of things to a lot of different people. For some people, it’s supply chain. For others, it’s accounting. For others, it’s like the system that does everything, you know, truly all in one.
For us, it really is a little bit more simple. Where you can have a hub spot for your sales force for your sales and marketing teams and you can have a QuickBooks or an accounting solution for your accounting teams, these operations teams and supply chain teams don’t really have that system of record to use from. And so when we say ERP, that’s really where we want to focus. The system of progress really comes from my consulting background. So I love consulting, being boots on the ground with these companies, solving these real problems.
And I found that as I was going and consulting for different size companies in different industries, they were dealing with the same problems. Didn’t matter if they were pharmaceutical, chemical, they were dealing with physical supply chain issues, planning, production scheduling, all of these things, where a lot of what I assumed to be true where all of these industries were so different was not actually what I saw in practice. And so when I said, hey, we can build a more modern system that works across all of these, I think that’s great and that’s needed.
That’s where then the consulting comes in and it’s like the value that I was adding was going in and saying company to company basically the same things over and over again. And during this time I was also reading everything I could get my hand on. So I read hundreds of books all nonfiction. My wife was trying to get me to read more fiction. I didn’t read a single fiction book. It was all nonfiction but no romance novels. No, not.
Beau Hamilton (12:27.709)
No romance novels or anything like that? No? Okay, okay.
Dan Koukol (12:33.972)
I really just became obsessive with this and I found that when you look at what’s actually moving the needle for companies at the end of the day, it involves educating these companies. If you take manufacturing company A and you have two versions, you have the ideal state version where this version of that company is doing everything possible versus kind of the status quo version, the first one is going to outperform on financial metrics, success, customer service quality, all of that.
And what that means practically is that a lot of these companies are run by really talented, really smart teams, but they’re stuck in the whirlwind. And so there’s just a lot of things that they’re not doing that they either know they’re not doing and intend to start doing over the next quarter or year, but just don’t get to it, or they don’t know they’re not doing. So if you say, Hey, what is our downtime? What’s the data? We have nine types of downtime. Can we sort it from high to low to figure out how we can improve our operations efficiency? Where are we losing margin? Where can we increase throughput?
When I was consulting through the books I was reading and through my experience, there are key things that every company can do to solve those questions. so system of progress is built on that idea where number one, you have to have a platform where you can get everything in. That’s important because that gets data into the system. You then use that data to basically take, you know, the best consultant in the world, which are foundational models, large language models, and you sit that on top of the data and you use that to proactively tell these users and educate them.
The best version of your company is going to do all these things. Here’s what you’re doing. Here’s what you’re not doing. The things that you’re not doing. Here’s how to do it. Here’s how to actually start doing preventive maintenance. Here’s actually how to schedule production line. Here’s how to optimize working capital so you can have less cash tied up in inventory and more cash spent on sales. And then we’re also measuring the stuff you are doing. Just because you’re doing it doesn’t mean you’re necessarily doing it great. It might be red, yellow, green on the scorecard. And so for us, the thesis is get the scorecard.
get a system of record that can get the data in one place. How are you doing across all key areas? And then use AI, use the large language models, use the system of progress to proactively guide and push these companies to accomplish things that they have been putting off knowingly or unknowingly.
Beau Hamilton (14:46.697)
Gotcha. Yeah, I think from a high level, I like the system of progress term because it’s kind of self-explanatory. At the end of the day, it’s all about progress and deliverables, what you’re able to extrapolate from the records and from the marketing side of things. think that speaks volumes. And imagine it resonates well with the stakeholders that you have discussions with and then how you’re able to integrate and relate that to the traditional ERP.
you know, platforms you’re working with or competing against, you know, it just, I like it. think you’re, I think you’re definitely onto something there.
Dan Koukol (15:24.152)
Yeah, yeah, and we agree. And I think and I think onto that note, like we don’t consider ourselves selling software. We’re selling progress. You know, my consulting experience, I helped these companies move that like I was obsessed at moving the needle. There’s a whole area of consultants that will charge you a lot of money not move the needle. I was obsessed at moving that. And then I became the CEO of a disc golf injection molding company and we grew revenue 5x in 18 months. We improved profitability to 21%. We
we actually were able to make meaningful financial statement impact, culture impact throughout every area of the business because we were able to take the data, unify the entire company around the central nervous system, which was essentially the scorecard that I used to manage that business. And when you can get an organization seeing everything all in place and you can align every individual and every team to that, and then you can go and execute and then use that data, use that software, use that system to see how you’re executing.
That’s where you really make meaningful progress and where you actually allow a lot of these companies who have been wanting to grow for a long time or improve margins or enter into new categories. That’s where you can actually have a system, a software help unlock that for them.
Beau Hamilton (16:33.769)
Right. I want to get into that some more. know a lot of it ties back to that system and that software you have built under the hood. I know one of the kind of architectural and technical terms you guys use and the models you employ is this object first model, right? Can you explain what that is in the first place and then why that maybe matters for things like stability when a company’s operations are becoming more complex? I mean, there’s never like a constant
you know, state of flow, feel like you’re either you’re growing or you’re shrinking and there’s a lot of complexity. So how do you how does that model factor in and deal with that?
Dan Koukol (17:11.342)
Yeah, think on the surface, when you talk about building a product in this space, I think there are lot of folks out there who, for lack of better way to say it, overcomplicate what it means to innovate in this area. And so they’ll throw out a lot of really complex, we’re going to use all of these crazy configuration, workflow, crazy things. And I think the reality is actually just much simpler. And so for us, we go back to first principles thinking and we go back to
What are the actual screens? What are the actual things these companies are using to conduct business? Those are sales orders, purchase orders, inventory labels, shipments, manufacturing orders, work orders, customer records. So when you look at any of these, like look at a Shopify, look at a HubSpot, look at a Service Titan, you have these core things that the entire system needs to be built on. Now, take a sales order or a purchase order. These are objects that can get quite complex. You can have a blanket
purchase order, you can have a standard. You can have a number of different things where you have different light items on a purchase order. Maybe some have different ship dates, maybe some have different price rules. So these objects, when you look at them, if you understand the space deeply, which we do because we’ve been in this space for so long, and you understand how those work in chemical, food and bed, textile, automotive, all of the different 18 sub verticals of manufacturing, what you can develop is a unified object model. In other words, you can develop a sales order that works for all industries.
that can do all the things that a sales order needs to do. Same thing with the inventory, same thing with the shipments, and that gets pretty complex. But if you build a system from that first and you get those fundamentals correct, that unlocks a number of things downstream. Number one, in the short term, that unlocks a number of different companies that come to Digit and you can just onboard themselves because they can complete a sales order workflow. They might think they have a unique way of dealing with a certain order, but Digit, because we have the object model correct, they can do that in Digit.
So getting those correct first is key. So it’s not as sexy as maybe these complex, crazy things that people talking like, I don’t know what that means, but that sounds good. We want to get the fundamentals correct. And then that foundation builds onto our product strategy on top of that of layering the workflows, the AI on top of that, foundational objects.
Beau Hamilton (19:27.573)
which we’re definitely going to get into. I definitely have an AI question for you, so be prepared. But I’m curious, so along with those kind of outputs and progress and effects you’re looking to help clients achieve, I’m curious how that improved clarity helps a company’s financial health. Can you talk about how that improved, again, clarity just turns into these healthier margins, cash and growth and all those good things?
Dan Koukol (19:56.866)
Yeah, this is one of my favorite, favorite topics and I can talk for hours and hours. So I think, I think at a high level, starts number one with education. and this is what a lot of people miss. And so first you have to start with the map. Here’s all the things as a company you should and can be doing because a lot of people you would assume if you go and ask what’s your OEE, what’s your scrap rate, some companies know it and can tell you other companies are like, what is that? Why is that important? And so step number one for us is here’s all the different things. Here’s the map here is
you know, look at your, if we’re looking at financials, here’s your profit and loss, here’s your balance sheet, here’s how much you’re carrying in cash and inventory, here’s what your gross margin and cost of goods sold are. And it’s starting with, here are all the different levers in your business that you can use to drive that financial performance, those customer service levels, those things. Then you start backing into the actual current state for these companies. So now that you know everything you should be doing, where are you today? What is your OEE?
What is your throughput rate? What are your margins? What does your pricing look like? And then it’s figuring out what is it that you’re not doing that can then improve each of those levers. So for example, you might go into an injection molding company and see that their cycle times to run a part might be 80 seconds, but it should actually be 65 seconds. And so while that 15 second reduction in cycle time doesn’t sound like a lot, when you look at the scale that these companies are running at,
That reduces cost of sale dramatically. That allows you to make more parts per machine per shift. So that increases your capacity. And that has very real impacts on cost of goods sold and ultimately profitability. When you look at, we’re not building an accounting system of record, but we do feed accounting systems and budget data. When you look at a sales order and you look at a customer that has terms and you look at managing accounts receivable, which some of these companies just frankly don’t do a very good job with, and you say, hey,
We have 30 day terms with some of our customers. Let’s send them an email response seven days before terms are due to increase the receivable rates so we can manage that so we can get cash in quicker so that we can then use that cash to spend in other areas of the business. And so it’s education in terms of here’s all the stuff you should be doing. It’s getting the current state, getting the data and you get the data from the system of record because you can manage inventory, production, planning, procurement, sales, shipping, all of that. Just so have data for all of those things.
Dan Koukol (22:14.87)
And then it’s driving these companies to figure out for each layer, how do you improve pricing to get more revenue? How do you improve production scheduling and planning so that you don’t run out of raw material, so you don’t shut your production line down? It’s dozens and dozens of things that you do where these daily actions compounded over time drive meaningfully, really strong improvements, both financially, but also at the customer side, internally as well from a culture side, so many things there.
Beau Hamilton (22:44.233)
I like how you mentioned education as like the first pillar, because I feel like that’s something that’s kind of glossed over when I hear some of the explanations and the software space where I like to call it the education, the learning curve, right? And it’s such an important factor and you almost have to have, it’s a two-way street where you have to kind of meet the client a little bit halfway and they have to be open to learning about the…
various processes, but obviously you have to kind of be able to translate and make them, allow them to absorb it and learn it more fluidly, I guess. But I think it’s the, ultimately, it’s the operational discipline, maybe that you could say, refer to it as that ultimately shows up in the balance sheet that you’re talking about. And then another thing I was thinking about is the timeline. So the in-between time it takes to actually see the results is sort of another advantage.
I know that Digit emphasizes being deployable in maybe days or weeks as opposed to months or years. How is that even possible in the first place? And what does that model do for a customer’s business risk?
Dan Koukol (23:56.856)
Yeah, this is one thing I think we demystify the most in the space. So, you know, why is it that these companies can stand up a complex e-comm store in Shopify in a couple days? Why can they get a CRM like a HubSpot launched in a couple days? But for some reason, you can’t do that for the upside. That’s always been a crazy thing for us because I think people are so grounded in ERP manufacturing software must have a three, six, nine month implementation because that’s just what’s been fed into, you know, the system over all of these years.
And so I think for us, like, you know, like why does Ecom, like why doesn’t, you know, it take you six months to do an Ecom store? Like you’re managing the same things, you’re doing the same things. And so it starts all the way back to that, that object level strategy we were talking about a minute ago. When we talk about, not doing custom software development, we’re not doing customization, we’re doing configuration. And in fact, we’re not doing configuration. You as the end user are doing the configuration. You know that you have batch production. So you’re going to use our production module to do
batch production. You have continuous production, so you’re going to use that same module to do continuous. it’s really the first step is if you have the right objects that work across industry, where a user can just sign up for Digit and create a sales order, easy. Everybody can do that in matter of a minute, right? Even if you’ve never used Digit before. And then you see a button that says make and you can make a manufacturer when we’re tied to that. And you just fill out the form and you create it. And then you start adding output from that. like it’s a very intuitive
flow that a lot of these other B2B SaaS apps follow, that should absolutely have been the case since day one, whereas all of these other users are conditioned to saying, hey, we need to go through six month sales cycle and implementation team is going to come in. We’re going to have all these three hour meetings. We’re going to do a launch. It’s going to be a 12 week implementation. For us, that’s crazy. It’s like, look, free access to the product, sign up, do it yourself. We have a great knowledge base that can walk you through all these things. But it’s really the configurable objects that unlock all of that.
And then it’s taking some of the best practices from other B2B SaaS apps that these technical or non-technical users are familiar with if they’ve used a CRM, if they’ve used a Shopify before, and being able to do that same thing on the production side with Indigit.
Beau Hamilton (26:06.783)
Yeah, that speed advantage really shows up when you integrate with those other services you kind of mentioned, right? You can’t try to replace everything. You kind of have to obviously partner with some of the kind of established tools of the trade. QuickBooks comes to mind. your sole focus is on one thing and doing it well, really well, you should try and work with it as opposed to reinvent the wheel. So we’ll talk about kind of the why unbundling
Dan Koukol (26:11.118)
That’s right.
Beau Hamilton (26:34.719)
the old monolithic ERP stack is the best way forward for customers in this space.
Dan Koukol (26:40.44)
Yeah, so we talked about systems of record early on and those are essentially platforms that help different functional areas in a business run. So, know, HR have HR systems, Gusto, Rippling, a number of these have come onto the scene recently. Accounting teams are using Xero, QuickBooks Online, QuickBooks Desktop. You have marketing and sales teams using the HubSpot. So companies are very familiar with over the last handful of years, these kind of broad systems that manage a function.
And so the biggest thing we’ve seen is there hasn’t really been a system of record for supply chain. And we talk to customers all the time that are looking, they’ve gone out, they’ve searched and they either are saying, hey, I can piece together all these point solutions for this or I can go try to implement a NetSuite or an SAP, but it seems like overkill. It’s also really expensive. It’s going to take forever. And so they’re basically looking for that final piece in their puzzle, in their tech stack puzzle, whether they realize it or not.
The CRM’s work well, the accounting GL works well, but nothing’s really working well for this final piece. So I think the unbundling has happened just naturally. And I think what’s happened is when you look at the sales, the accounting, the HR, they had point solutions at the very beginning as well. But those point solutions ended up becoming systems of record for those categories. And so I think operations has been a little late to the game. Manufacturing typically is historically, if you look at that over time.
And so I think the next step is that, I think, you know, 10 plus years out in the future, I think you could have these systems of record ultimately converge into the true, you know, single systems to rule them all. But I think that natural unbundling has happened just, just over time naturally as the point solutions and different functions have consolidated.
Beau Hamilton (28:16.148)
I like that framing and that explanation. think that makes a lot of sense. You don’t want to force businesses into one kind of lockdown service. It’s important to have an unbundled system that just plays nicely with the other tools out there. It’s more user friendly. Because everyone’s using different tools and integrations, So it just makes the learning curve more challenging if you’re trying to, again, reinvent the wheel with your own solution.
So we’ve made it this far into the interview without really talking about AI. We’ve kind of teased it here and there, but it’s a big part of kind of the conversation on a macro level. I’m curious, where do you see AI delivering real value first in this space? Is it more as maybe a coach or something more as like an autonomous worker or an agent, which we kind of here talked about in the industry.
Dan Koukol (29:08.258)
Yeah, we’re a little contrarian here and we’re a little bit more pragmatic. So, so we firmly believe the short term value is really going to be coach. Now we think the agentic thing is, very powerful and we’re not discounting that, but we think in the sequence of events, when you talk about getting AI to be adopted heavily by a broad user base, the coach is where you win. The reasons for that are think about what, what AI is really mostly now it’s the foundational models. It’s the LLMs. It’s the predicting the next word in a sequence when you peel back.
the layers. you know, OpenAI had a paper in terms of how people are using ChachiBT. A lot of it is knowledge, education, think about how we all use it every day, right? We pull up ChachiBT, we’re getting it to explain something, we’re getting it to kind of go broader. And so, LLMs today in their current state with the models are the best at sitting on top of broad swathes of information, knowledge. They’ve read more books than I ever can, right? They are the best consultants in the world.
It’s taking that and it’s unlocking. It’s going back to that point I made about education earlier and it’s, it’s unlocking that, that first level immediate education. And it’s teaching these operators, these managers, these owners, how to think about their business in a more complete and holistic way. And then it’s allowing them to use the data. LLMs are very good at using structured and unstructured data, which we provide plenty of being a system of record at the core. And it’s taking that and it’s saying, look, two weeks after using digit,
we can already start using that data to start bringing to your attention things that might be able to help you increase revenue or increase margin or get the organization to be a little bit more aligned around a one page scorecard that you’ve never been able to actually have as a company because you had a bunch of point solutions or you had pen and paper and you weren’t able to roll it in. And so that is the, that one that mirrors how these companies are using a lot of these LLMs today. And two, it’s a much easier technical lift for us to just go and launch.
Agents, I think, are really important, but there’s still a lot of work to be done. We think it’ll improve quickly, but when we talk to people, there’s still a lot of human in the loop that’s going to be required for these things. Management and a lot of the companies we’re talking to, they’re just not ready for it. They’re not ready for it, even if you had them there. On the maturity curve of first, you need those agents to sit on a really powerful foundational system of record that has the data, that has that completeness, that has that muscle memory. Memory is a big part of the AI strategy as well.
Dan Koukol (31:34.304)
And then you can start layering on the second or third order effects, which is more of the replacement layer. So first it’s coach, get the education there, get these companies to think about their business in a different way, in a complete way, get them to look at all of their data and then based on their strategy or what they want to do in the next quarter, the next year, the next five years, they can decide what they want to focus on and then they can learn what is predictive maintenance, why is it important, why should I have a plan for this.
Why is managing working capital like ask a bunch of people what working capital is and a lot of won’t be able to clearly articulate why that’s important Now they start becoming smarter now they start having consultant the best world-class consultant built into the software themselves And every day when they log in they’re learning learning learning and that’s where we think the value comes in in short term
Beau Hamilton (32:16.787)
I think that’s a really real transparent answer. And I think a lot of it comes down also to, of course, the timeline will eventually get to this almost dystopian future where AI is doing all the things and humans can be drinking Mai Tais on the beach. But that’s still quite a ways off, right? And until then, humans are going to be running the show, and AI will help guide and better extrapolate the data and lead to efficiency gains and whatnot. And I think it’s also kind of a
It’s just a byproduct of our kind of economy and society where we’re always chasing the next shiny thing. But let’s just start at the basics. like when AI kind of first made a splash in the scene from on the consumer level, it was just bringing us all this data, right? It scraped all this. It read all these books, scraped all this data, became a really great resource. And then now it’s all the talk is like how it can work for us.
as a, a worker standpoint, but let’s not forget kind of the basic main functionality and it’s just bringing us data more efficiently. So I think, I think you got the, your focus is at the right place, at least from my perspective. Now, speaking of kind of guidance and the coaching, you talk about companies having a hidden blueprint of everything they should be doing. How does making that operational roadmap visible, align teams and help them kind of stop relying on what’s a
referred to as tribal knowledge. know? Tribal knowledge.
Dan Koukol (33:44.59)
Yeah, tribal knowledge we see. I think every organization has tribal knowledge to a certain degree. I think, again, going back to the basics, is the best companies are the ones that are most aligned. And alignment means a lot of different things. So at a core level, alignment means you have all of your data, you know what you’re doing, and every individual and every team, when they come into work every single day, they know what they need to do in order to move the organization towards those key set of objectives and key results that you’ve defined.
or that leadership has defined. A lot of what prevents companies from becoming aligned are these pockets of travel knowledge. And these can be very technical, very specific. We used injection molding before. can bring it up again where there’s a number of different settings, pressures, heats, toggles that you can put into an injection molding run where you might have a shift leader go and program all those machines where the operators don’t know anything.
and you don’t have that knowledge shared or documented in the organization. You also might have somebody who’s doing sales order entry and certain customers maybe are easier to deal with than others or you might have certain pricing information where you could have done one thing or another. And so for a company to be fully aligned, it’s getting a lot of that tribal knowledge out of the individual’s heads and it’s getting them into the system and that’s done with the data, right? Capture all of the injection molding data, capture all the data that’s used.
to manage customers, manage items, to manage follow-ups, to manage everything. When you have that and you combine that with, again, the central nervous system of the scorecard, and everybody knows what gets measured gets managed, right? That old adage. So when you’re measuring everything and you know how long it takes to ship something, you know how much money you’re making on those shipments, you know how long it takes to run a finished good through production.
you now have all of the teams and individuals understanding, you know, here are the different projects or initiatives that I can do to influence my metric, my team’s metric. And then when you do that on a daily and weekly basis, that’s where you these companies that are starting to out-compete and outperform their peers. And so that hidden blueprint now becomes a blueprint that’s centered in the scorecard, that’s centered in the data, and that’s centered in the execution that is really built on the back of the proactive guidance that the system of progress is providing.
Beau Hamilton (36:06.473)
Makes sense. It kind of goes back to that education pillar that you’re talking about and just getting, just kind of explaining and showing people the light and what’s out there and the other options out there and getting them to kind of take the leap maybe and give a different kind of solution a try, get out of their bubble, so to speak. So I want to circle back to the point about how many mid-sized companies still rely on the dreaded spreadsheets.
I believe I mentioned that in the intro. Why do so many of these companies stay there for so long, and prevent themselves from adopting some of these more modern tools and solutions?
Dan Koukol (36:36.835)
Yes.
Dan Koukol (36:44.802)
Yeah, I think ultimately it’s because they in a very odd way have the most flexible object model still. So spreadsheets, they’re their biggest pros that they’re flexible. They’re easy to understand. Most people have used Google Sheets or Airtable or Excel at one point or another. And so where the ERPs were too rigid and a sales order didn’t allow you to do something or you couldn’t track certain data in a manufacturing process or you can do a certain report, you’re able to do very quickly as a non-technical user in Excel.
You can bring the data and you can create your own rows and you can do all of these different things. And so what you have is the operators, the users, the admins fall back to spreadsheets because they’re the best tool to fill the gaps left over by legacy systems. And, know, we’ve seen, we’ve seen extreme examples of this. We’ve seen five year SAP implementations not work well where 17 of the 20 something core workflows ended up back in Excel after millions of dollars to spend. Right. And, and,
I think every company in any industry, any size probably resonates with that. They’ve probably seen that to some degree. And so for us, again, that’s why we place so much important on the object model. If you can get them off of the spreadsheets and into the tables and forms that allow them to capture this data more automatically and in a simpler way, and you expose that through the reports that we have in the system, through table functionality where we mimic a lot of table functionality in Digit.
launching things like keyboard navigation, inline cell editing, grouping and filtering, pivot tables in the app, all those kind of things. That’s where you finally start to wrestle these companies out of Excel into an actual B2B SaaS product. And so it’s not that these companies want to use Excel. Most of them don’t. mean, Excel hell is a term for a reason, right? That doesn’t evoke feelings of positivity. And so I think when you finally have an application…
Beau Hamilton (38:28.361)
Yep.
Dan Koukol (38:34.778)
that brings in some of these things. And you look at the air tables, look at the smart sheets, look at the notions even to some degree, but you can attach them to ERP specific primitives. That’s where you finally get this aha, this unlock. I can create my own type of table. I can attach it to my own type of form. I can connect that to inventory logic and allocation logic and sales order fulfillment logic. That’s where you start to get really a lot of that magic where no longer these spreadsheets are needed to fill in those gaps left over by the legacy systems.
Beau Hamilton (39:02.941)
Yeah, it’s taking some of the best features of Excel and integrating them into your platform and also kind of making it sort of welcoming to the Excel users out there who are maybe burnt out in Excel hell. All this talk has me thinking about the future and where things are headed. What do you see as the single most powerful structural advantage, whether it’s data, maybe hardware integration, you think of network effects. What’s the most powerful structural advantage that will actually determine the leader?
Dan Koukol (39:10.19)
That’s right.
Beau Hamilton (39:31.033)
in operational software over let’s say the next decade. I mean, if we can even think that far in advance, you know.
Dan Koukol (39:37.806)
Yeah, I think ultimately it’s two flows. It’s product and distribution at the end of the day. I think on the product side, it is multi-layered. I think data is huge. Data is on the proprietary data that you capture both on the software and the hardware side. So think IoTs, think sensors, think all of these things, even robotics to a large degree. If you’re able to plug that into the nervous system of your product, you’re going to win because you’re going to capture more data.
Network effects are also very real. Supply chain is very interconnected. Any one of our customers today has a whole bench of customers and suppliers that would benefit from using digits. You look at a lot of the waste that still happens in supply chain, the manual sending purchase orders and sales orders between organizations, the lack of inventory visibility between organizations. There’s a lot of speed you can introduce into the supply chain when you automate those communications and data sharing between companies.
So I think I think the company that unlocks that the best is going to win I think ultimately the company that has the most beautiful product that is the most flexible Is going to have and create for the first time in this this segment of the industry irrational product loyalty You know think about the the Apple fan base the iPhone like you know back in the day if you can create that with a beautiful UI That is really flexible and powerful. That’s huge and then distributions the other side. So we we subscribe very heavily to
the narrative out there where if you can create your own category like system of progress, for example, and if you can educate the market on that, whoever can create and own those categories typically is crowned the winner of that category. And historically, I think the number is they win on average 76 % of the total market as the category leader. So for us it is we have to nail product that is probably the most important, but then we have to be the most vocal. We have to be the ones creating this category.
And when we do that, when people think of system of progress and they think about what scorecard means and education and guidance, they’ll think about digit and they’ll associate that new era of software. It’s no longer ERP, it’s system of progress. They’ll think digit. And so I think the companies that win have to be the best at both product and distribution.
Beau Hamilton (41:47.253)
Very well said. I would have to agree with you. you’ve given us a lot. It can be a lot to think about kind of where things are headed. And there are some kind of the buzzwords and keywords that you mentioned. think robotics is definitely the next kind of big frontier with manufacturing and where some of these automated tools are headed. And then obviously how your system integrates and allows people to kind of unlock the next big thing I think is going to be something to look forward to.
I want to give you the chance to give us one more big takeaway message for listeners who made it to the end of this episode. If you could share one maybe foundational truth about, let’s relate this back to the physical economy, where if you can give us one truth about the physical economy and where it’s headed, what would that be? What should leaders really take home and maybe eternalize for themselves?
Dan Koukol (42:41.79)
I think plain and simple, a new industrial era is beginning. And this is widespread. This is obviously AI turning systems of record into proactive systems of progress. It is combining hardware and getting the data from those sensor technologies from physical real world operations automatically captured in these systems. And it’s the use of robotics to automate a lot of things that haven’t been automated today. I think…
That at the core is something people have talked about for a long time, but over the next two, five, 10 years, you’re going to see that adopted at a much larger pace. You’re going to see costs for robotics and hardware start to trend down, be more accessible to a lot of other companies. And the folks that can integrate that all together in a single system, I think are going to unlock a tremendous amount of value to think about the headwinds, right? Think about, the tailwinds rather, think about the ability to get these systems launched much quicker than at any point in history.
they’re going to have much more coverage, both software and hardware side, getting the data in. Think about tying that into the AI models that can take that data and highlight what it is you need to do. And think about the agentic piece that can automate a lot of that stuff for you over time as well. So I think you’re going to see this really rapid increase in, know, we expect GDP at a national level and global level to start increasing simply because there’s still so much unbelievable waste in the physical supply chain that the hardware, that the robotics, that the AI is going to eliminate over time.
And I think the way that these companies are viewed and managed is also going to change. I think you’re going to see a lot more scorecards, a lot more alignment through organizations. And I think that’s going to be a big win for the physical economy.
Beau Hamilton (44:15.733)
Amazing. All right. Well, for those interested in learning more about Digit and getting in contact with you and your team to stay up to date with a lot of these insights that you just mentioned, where should they go?
Dan Koukol (44:26.466)
Yep, go to our website. is www.digit-software.com.
Beau Hamilton (44:31.081)
digit-software.com. All right. Thank you so much. That’s Dan Koukol, CEO of Digit Software. Dan, it’s really been a pleasure. Thanks for sharing everything that you shared with us. We appreciate it. Thank you all for listening to the SourceForge Podcast. I’m your host, Beau Hamilton. Make sure to subscribe to stay up to date with all of our upcoming B2B software related podcasts. I will talk to you in the next one.
Dan Koukol (44:42.466)
Likewise, thanks so much, Beau.