Revolutionizing ERP with AI: What You Need to Know

Full transcript

AI changes everything

Welcome back to the podcast, guys. Today, we are joined by Don Waters, the principal and tech evangelist at BC Computers. Don, welcome to the podcast. Well, thank you, Nicholas, and it's really good to be here. Thanks for having me on. Why don't we start by you telling us a little bit about yourself, Don, and what are you guys actually doing? Sure. So we are a solutions provider, cloud based solutions provider to primarily businesses in what is called the ERP arena, enterprise resource planning arena. So what that basically is is just business management systems, we tailor them to small to medium sized businesses, and what we have done for many many years is we kinda take tool sets that are usually only available in the higher end enterprise and we bring some of those tool sets down for the small to medium sized companies to be able to take advantage of those. So that's the primary emphasis of our business, and a lot of it is done with manufacturers. We specialize in manufacturing. Mhmm. I saw that you were a lab technician.

Yeah. Yeah. So I started out as a you know, I was a chemistry major in college. And I came out of Rutgers University here in New Jersey and I was working for a textile chemical company. And what happened was I got involved in spectrophotometry, which was testing dye stuffs and colorants for the textile industry, and I was very intrigued by the computer that I used doing that. And as it ended up, I said, you know what? I'm not so much into the chemistry, but I sure like the computers. So Mhmm. That's that's how that happened. And then I was joined by my brother who was a University of Texas graduate in computer engineering. So we put our heads together and started building software for chemical manufacturers, compounders, and blenders. That's how we started in the business. Interesting. You've been in the business for over forty years, if I'm not mistaken. Yeah. I think it's about forty two right around now. Yep. Mhmm. What are kind of the trends that you're observing in the ERP industry as of now?

Yeah. Sure. So like everybody else, you know, AI is the buzzword, but AI is also the reality. You know? So if you get past the buzzword and you look at the realities of what it's doing, it's really changed the landscape in the last two years and more to come to say the least. Because what's happening is this, it's not just so much that you have AI being incorporated into the ERP platform, but it's more so that there are tool sets that are available. Everybody has other applications they use. Everybody has other needs for integrations and automations. These AI tool sets are providing a whole new landscape for that to occur. If I can, I'll explain a little bit further what I mean by that. It used to be when you looked at ERP, you looked at it and you went, okay, we can't find exactly what we're looking for, should we build it? And you look at what it cost to build it and it was a fortune, a lot of money. Small companies couldn't do that. So you decided to buy it as a monolithic ERP system because building it was not viable financially.

Automate everything

Right? Now different playing field. You can buy something that doesn't do all of what you want it to do and you can create all kinds of automations and integrations around it to get the rest of what you need and the price point for doing that has dropped, you know, logarithmically over the past few years because of the injection into of AI into the all the tool sets that you do this with. So we spent years integrating ERP with other applications sort of the hard way. And now we're able to do that in a much easier way and therefore less expensive so that we can bring the tool sets the big boys are using down to our SMB clients. Mhmm. I can imagine that it was very hard to do this. It isn't you know, people always see the end product. So do you have some interesting technical challenge that you had to face when building this? Yeah. So what it really is is learning the extent and, you know, every tool set has its its personality, if I could say it that way. There's there's what's called the design build philosophy behind all software.

Right? How we build it and how we design it is how you're gonna end up using it. Well, you have to learn what that is and so we had to vet all kinds of tool sets to figure out which ones can do what we need to do, how difficult are they for our engineers and technicians to learn, what do they cost, And are we truly gonna be able to use these on an ongoing basis to provide integration and automation services to our clients already on ERP platforms? Mhmm. Could you describe any specific AI tools that you are using or kind of the use cases there? Yeah. Yeah. So we're big into n eight n. Alright? N eight n comes out of UK, but we love the product. We're working very heavily with it. It does what we need to do mainly because, you know, there's all these low code, no code, right? But the issue there is when you hit something you can't do with the low code, no code, you're stuck. And we have too many circumstances over the years that we've already seen, you know.

I need to go out, get data from over here, but then what I wanna do is take that data, go out do some research on that data from AI. Bring the result of the research back in, add that to the data that I already collected and bring that into my system to present it to the users. You may have to do that with three, four, five, six, eight steps, and that was impossible to do before AI. It was just totally impossible to do. So we get challenges all the time from our clients that are very sophisticated. So we wanted a tool set that isn't like, okay, great, it's low code. I get that. But what else can I do with it when the low code stops? Right? When the low code hits the wall, where do I go with this? And n a n is completely JSON based. Right? And that gives us extended tool sets to work with so that we can provide the solution for the little guy with something simple or the, you know, the bigger guy who's got a very sophisticated automation that he's looking for. Mhmm.

When low code fails

Would you think of agentic AI and the whole bubble around OpenCloud and kind of the autonomous AI agents? Yeah. So I think like anything else, you know, the toolset is the toolset. You still need human eyes on it. Don't ever do anything terribly terribly important without putting human eyes on it through the process. And particularly, if there's finance involved or there's risk involved, you better be looking at it before you put it out to anyone anywhere. And so I would just say to people, caution is important because AI can give you incorrect information, and it does. So you have to you have to, you know, do do limit your risk to that for sure. Mhmm. When you are making the architecture and the design decisions, how do you approach it? Well, what we do is this. We always sit down with our clients, and the first thing we do is we look at a workflow. What are you doing now? What are you looking to do that you can't do? Right? And we sit with them, we go through that in great detail.

Okay. So tell me how you do it here, what would that look like there, what if we could do this? So that we come up with at least a preliminary specification for what that automation or integration is gonna look like. Right? And then we sit down and we whiteboard that, if you will, into a workflow on a diagram so that we have something to control for the engineers to look at, And then they sit there and they go and say, okay, here's how we can do this. And then they come back and they ask questions. So it's an iterative process, but basically starts the design in the beginning has to be very clear so that you have a clear path to follow. But sometimes what happens is they come back to say, oh, hold on a second. We forgot. There was something we didn't tell you here. Alrighty. Great. We just go in and drop another node. Right? We drop another node in between the two that were already there and add that new function in that node. You know?

So it's all about getting the proper design up front, working that through, bringing it back to the client, getting their viewpoint on it. Is there anything we're missing here? Is there anything we didn't do the way that you specified it to us? No. We're good. Or, oh, you know, we forgot, which happens almost all the time. We forgot something else. Can you add this in? And then we go back through, do the same thing. We go back to them with the result, and then they sign off on it. Don, what's your vision for BC Computers over the next few years? Yeah. So what we are heading towards, which was not at all possible, you know, dollar wise, is we're heading towards what we call hybrid ERP, some might call it composable ERP. And what that basically means is you have other tool sets that you use and they're important to your business, but what you want it to look like to your users is it's all one in the same. So the users have no idea that they're off and AI is doing this work for them,

they just know that what I what I want I can get in the form that I need it and AI can help me do that. But beyond that, AI also gives the user the capability to, hey, I wanna speak a report into the system. Give me q one sales distribution companies only, product category a over $10,000. Tell me how much I sold of that and AI will go create the report for you, It'll show you the report. Is the format okay? Would you like to change the format? No. I'd like to change the format, move this to there, that to here, etcetera. Okay. Great. I have this report. Do you want me to automate this for you? Yes. Automate it for me. Great. When do you want this report run? How do you want it delivered? Well, do it for me every Monday morning, deliver it to everybody in c c suite level in the company. That kind of stuff. Mhmm. If there is someone that would like in to to get into your industry, what would you advise them if they would start this year? Yeah. So in our industry, the main

thing is, you know, you can have a lot of AI skill set, you can have a lot of software capability. You have to understand the business. And so what I would say is is concentrated isn't just about the tools, it's about the relationships you build with the people in your business and the way to build those relationships is through trust. And trust is only gained when people understand that you know what you are doing. You understand their needs, you understand their problems, you understand their pain. The only way this really works is when that's the basis to begin the relationship with. Everything else is just layers on top of that. So the tool sets, the AI, the composable ERP isn't gonna happen for anybody unless they trust who you are and they know that you know what you're talking about. So you have to and that's a knowledge set. That's a knowledge set that a lot of people don't have because they've not been in the manufacturing space or distribution or service which are the other areas that we serve as well.

Thank you, Don, for joining us on today's podcast. I will make sure to add links so they can check you out, and thank you guys for watching. Great. Thanks so much, Nicholas. Appreciate it. Take care now.