[Podcast] Within Tolerance Podcast Episode 46 - Sunny Han and the Future of ERP for Manufacturing
Fulcrum's CEO, Sunny Han, joined the Within Tolerance podcast for a conversation about his journey into manufacturing and the creation of Fulcrum Pro, an innovative ERP software. Han's passion for manufacturing began during his college years when he decided to focus on business, particularly small businesses. He gained valuable experience working with Fortune 500 manufacturers in Minnesota and later switched to small business consulting.
Han's dedication to working with small businesses comes from the satisfaction of seeing the impact of his ideas and work on these companies. His transition from consulting to creating Fulcrum Pro was motivated by his desire to create a more meaningful change in the world of manufacturing. The tragic loss of a childhood friend made Han realize the importance of making a difference in the world.
Fulcrum Pro, an ERP system designed by Han, aims to revolutionize the way manufacturers operate, helping them improve efficiency and productivity. By combining his expertise in coding, passion for manufacturing, and commitment to helping small businesses, Han and Fulcrum Pro are making strides to change the manufacturing landscape for the better.
Listen to this episode, and follow the Within Tolerance podcast on Instagram.
Transcript:
Dylan Jackson: All right, guys. Welcome back. Another episode of within tolerance, this is episode 46 and we are joined with Sunny Han from Fulcrum Pro, which is a really interesting software ERP system. So I'm just going to start it off how we start every other guest off. So, Sunny, how did you get into manufacturing and then we'll lead into your company and your software?
Sunny Han: I learned to code when I was a child, my mom was a graduate student at the University of Minnesota. She taught me how to code. I made some video games, started building computers for friends and was really into math and nerdy things. But, in college I kind of had a revelation that I wanted to do something that would change something fundamental about how our economy works and decided to go into business and specifically small business. So during the recession, 2007, 2008 is when I started my career. And over the course of eight years, worked with roughly 350 different manufacturers implementing ERP systems, helping them select them, customizing them, and writing custom software on top of them. I met a lot of really great people and learned a lot about manufacturing and just got a huge passion for it.
It always bothered me how the system that we live in works about how people don't really understand how the goods that they use and are surrounded by work. And I was always that weird, annoying guy talking about giant flanges being used by nuclear power plants and how they're, how they're milled and in what kind of, what kind of steel and what forges they come from. But that's the stuff that I really geek out on is that, you know, since the dawn of time adding energy and expertise to materials, to make them more useful is how we've grown to where we are now. And now we seem to kind of lost a little bit of love for that and I'm trying to rectify that in any way possible.
Dylan Jackson: So to dive a little more deep, did you start out as your own consulting company, or did you start working with one of the bigger names or a reseller or, you know, how did you, what, what was your first break into that?
Sunny Han: I worked with a lot of bigger companies first. So most of my original experience with Fortune 500 manufacturers here in Minnesota. There's a lot of manufacturing here. We're close to companies like Caterpillar, Polaris, 3M, so there's a lot of opportunities to work inside that network. I got a little frustrated with how politics work in bigger businesses. I took some time off, took a couple of weeks off and worked for a friend's cabinet shop. I helped them tune their operations and they were so open to any sort of advice, any sort of, they were a lot more, a lot less meeting driven, a lot more experimental and just do something and see if it works. And we changed that shop around. We went from losing $10,000 a month to making $40,000 a month and just six months of me working like weekends and a couple of, of really short spurts here and there that I think was like the first hook was, the difference between bigger businesses and smaller businesses that smaller businesses typically are much less political, much more results focused, and being able to see your ideas and your work turn into actual results and outcomes, was really addictive. So that's where the commitment to continue to work for small businesses came from. And that was kind of the first experience in switching from bigger business consulting.
Dylan Jackson: Gotcha. So I guess that gets into your company now, which is Fulcrum Pro, how did you go from, you know, consulting with all these companies and then, you know, bouncing around some big manufacturers to deciding, to write your own or, you know, create a business around your own ERP?
Sunny Han: That's a great question. I had a comfortable life. I had a few people that I was managing. I was very in control of my own schedule. I was well-regarded in my company. The clients that I was working with were really happy with my work and I had a friend in high school, early high school when I was a freshman in high school. He was my best friend and he died of carbon monoxide poisoning at a Tomahawk Boy Scout camp. He was one of the only kids left over. He was helping close things down and, there was an accident and there was no fault of his or anybody else's. And it was, it was a really tragic moment. And I got a letter from his brother out of the blue and he basically just asked me like, Hey, you know, we, we talked before at his funeral about where we would be, and I just wanted to know what you do with your life.
And if you think that he'd be proud of where you are. I know this is probably a little too weird and soft and squishy, but that was really the moment I realized that all the passion and energy that I had to change something just wasn't wasn't there. I was comfortable and not really putting myself out there not doing anything to really change the way that things worked. It took about two months to close everything up and I handed in my resignation and started the business basically six weeks after that. So that was the initial dive, but I'd always been the person who would have a couple of scotches and, and talk some about how I was going to do this and do that, but that was kind of the wake up call to push me over the edge.
Dylan Jackson: Well, that's great. So your company is called Fulcrum Pro, I guess let's get into, you know, what do you do? We've hinted that it's an ERP, but, you know, what do you do as far as what modules do you have all of that stuff? And then I guess before that, maybe like, that seems like a very daunting task to start blank slate. I mean, I think I'm sure as a co-designer, that's a excellent place to start too, because you're not dealing with anybody's leftover terrible code, but how did you approach building something like that? I mean, looking at your website and looking at your software, it's not like it doesn't look like something that was, you know, hashed together quickly. Like this is very clearly a very well thought out program with lots of R & D into it. Like where do you even start when you approach a task like that?
Sunny Han: Well, thank you for that. We were really proud of what we do, but we also are always trying to push to make things better. But when I first started, we had just started building a house. We were living in my parents' basement. My wife was not happy with not having my income for an unforeseeable amount of time. But like you said, I knew the space was going to be really complicated and I wanted to build something that wasn't just a tool, that would handle something that was really important. And so with my math background and all that implementation experience, what I realized was that there was a lot of really cool stuff going on with machine learning, with computational algorithms and the key to unlocking that potential was having more data. I'm sure you guys know in most shops, data is on paper.
We call that dark data. It's not analyzable, you can't do anything with it, right. And if it's sitting there and it's sitting there for two weeks, maybe you have, you know, four operations in this particular routing and it takes two weeks to get it all done. All that data is completely inaccessible for that time. Maybe you enter it in a week after it ships. Maybe you run a report a week after that. All of a sudden two weeks have passed by, and you're looking in the rear view mirror, and seeing that something went wrong and you have no idea how to fix it. So I really took a lot of time to dig really deeply into the problem. I met a really great, customer, Central Rubber, they're a rubber extruder. We built an entire ERP system on top of JobBOSS basically filling in the blanks.
We built an entire custom ERP for an antenna manufacturer. We built in an ERP platform end to end for a laser perforation company. And we selected these companies that were in very different industries and had very different structures to how they worked, because we wanted to make sure that whatever we built was going to be solid in core problems. Like you said, not inheriting the baggage of systems that were invented in the past. And that was really painful, it took years. It took years for us to make the same mistakes that, that Epicor, that, that Dynamics hass made and realize them and fix them and continue to do that research and development. And a lot of the times the difficult thing for anyone is to look at something you did that you're proud of and say, this isn't good enough, tear it apart and start over again.
And so what you see on our website, what you see if you use our application is lots and lots of painful ripping apart and putting back together until it's better, better, better, better. So that, that really is our method. We want to deploy a piece of software that doesn't take a half million dollars to implement that doesn't take 18 months, that you can use if you're a small business, or if you're a bigger business that has all the complicated gadgetry and can deploy that does auto scheduling really well. That helps you estimate your margins using complicated math. And it shouldn't feel like it's difficult. It shouldn't, there is no reason why something that's good and complicated must feel difficult. And, and that really is a philosophy that we started on. We weren't confident. We couldn't really see any examples in the marketplace that did complicated things but we saw other examples like Shopify and in other sectors where this type of work was done, we reached out to a few people, talk to people, got as many ideas as possible and, and really started with an end user first mentality and built everything from there. So it's taken, you know, five years now to build the platform to where it is. But we're really proud of what we've accomplished in that time.
Dylan Jackson: Sunny, let's back up a little bit for those that are listening in that don't know what ERP is. Talk a little bit, bit about that. You know, for example, myself and Dylan, we both are relatively small shops. You know, he's got a partner, and I'm just a one man shop with a couple machines. What could your software and system or ERP in general do? How could that help us? How can that eliminate problems and, you know, streamline our efficiencies?
Sunny Han: So the best way to think about an ERP is that it's a centralized data store that allows you to do really cool stuff, because that data is centralized. If you imagine we talk about this all the time, our philosophy is that you don't need any software to run your shop. You could, no matter how big you are with the right people, the right talent, paper, clipboard spreadsheets, whiteboards, you could manage everything. But imagine if all that information was in one place and imagine all the tasks that you can automate, if that was the case at its heart, that's what it's about is not entering things in multiple places is having all the data available for everybody simultaneously. And it's for having a single source of truth so that whenever you reference it, you can get the same answers. No matter who's asking the question, those are kind of the first core principles of what an ERP is supposed to do.
Now there's a lot of different philosophies and different products and different things that are out there, but as a whole that's how you should think about it. If you're thinking about it for your company, for us, we focus on the guts of everything. We want to make sure that the actual selling, the actual making of the product that you're shipping out, whatever service that you're doing, the actual reporting on it and managing of the business. That's what we really focus on. So we start at quoting. Really, we want you to feel like it's an extension of your brain. We imagine that everybody who's quoting is looking at a spreadsheet, drawing something else on a different monitor.
And we want it to feel frictionless. We want you to be able to come up with a bomb and routing, associate those routing steps with the bill of materials and, and do it with your keyboard and mouse in a graphical beautiful way, and immediately see what the roll-up of the cost and margin is. And have it tie into your real production data, knowing that maybe every single time that you run anything of this material type on this lathe, it's always 15% more than you expect. So whatever your standard operation times are, we're just going to add 15% to give you that buffer back. And so you're pricing things correctly. Maybe we want to know what our capacity is. Can this be, you know, can it share a setup? Maybe it has half the same tooling as something else. And we can use that to combine it, to get some edge and price a little bit lower, whatever that may be.
We want to make sure that you're selling at market value and understand what the margin really is. And so that's really where everything starts. We keep all those quotes, we have them easily searchable. It isn't the type of search where you have to type in different fields and click and wait, you just start typing. And it feels like Google, it starts giving you results. You can convert that into a sales order. And for us, a sales order is all about planning, who it's going to, who ordered it, what the due date is, where it's getting shipped to. Is this, does it have a blanket order associated with multiple releases, multiple shipping destinations. And from that point in time between the bill of materials and routing and the sales order and a shipping destinations, our purchasing engine can help suggest when you should buy stuff, what you have on the shelves that you can use, , when the schedule should happen, what order you should do it in and where it drops into the existing schedule.
We have a job tracker, that's not paper, it's a digital platform that can be deployed on any tablet. You enter the information directly in there. If a drawing updates on the sales order, it immediately appears you don't have to worry about having old drawings, the data's live. You can report on it live and see what's going on in your shop, your report, the scrap and the quantity made. And by the time it gets to shipping all that stuff is in there. , we have an understanding of NCRs and CAPAs, of ASN 100 and ISO 9,001 standards we can help keep you compliant in that way. Shipping has integrations with many LTL companies and FedEx and UPS, and once that's shipped, we send an invoice out and our system is kind of gone. We have inventory interfaces that help you do stock-take really well all in one nice mobile interface.
You just press a button, scan a barcode, whatever it may be. We can print barcodes, read barcodes, do stock adjustments, lot traceability, kind of everything that has to do with being a really cool and really well-run shop. We integrate with QuickBooks and Peachtree and SAP, and almost all the software platforms out there for accounting and we have a really great receivables interface and tables, interface, and inventory value interface, but the actual general ledger and all that stuff, we keep out of our systems. So we can focus on the manufacturing pieces, a bit of a long-winded answer, but hopefully that's pretty exhaustive.
Dylan Jackson: No, I think that that's, I mean, anybody who's listening to this who knows an ERP system, you know, JobBOSS to whatever those are exactly the kinds of things that they would want to know. So I guess that kind of leads into this question. You've definitely touched on this, but like, what does Fulcrum do that others don't? For some of our listeners who haven't used Epicor or JobBOSS, you know, what, what sets you guys apart?
Sunny Han: We want the new future technology to be something that is usable by smaller manufacturers. My experience with people that are using JobBOSS, E2, etc, you'd have to type in a job number in, if you don't know what to click a little search button, and that search gives you a whole list and you have to search it and then find it, or you type it in and press tab, you have nowhere to go to kind of see all your data. If you want to report, you have to go into a report engine that uses Crystal reports. You’d go into the server to kick somebody off if they're logged in. You have to kind of print out a ton of paper to put it in the shop. Even if you use the mobile data collection or the workstation driver, you’ve got to walk to a workstation and enter data.
And there's only a certain number of fields that are there. We don't want that experience. That's not what technology is about. Technology from when I was a kid geeking out about everything is being able to make you and everybody that you work with much more efficient. It's about doing more with less time. It isn't about automating people's jobs away. There's so much work that we can do that we can pull back from overseas. If we're a little bit more efficient, that's really kind of the core philosophy that we have. We want it to be paperless. We want the data to be live. We want to give our users unlimited seats. You could, anybody that's in your business can come into the platform. We want the interfaces to be really beautiful and delightful. They should feel like really well-designed mobile applications. They should feel like really well-designed software.
In general, you have QuickBooks Online, you have Mint, you have all these things that you're used to, and yet your ERP software looks like it was written in 1992, because it was, and so it might not seem like a big deal, but you've probably seen it. There are customers of ours that use Epicor and the end users, the, the person who's entering in the scrap quantities. The feeling that they have is one where they cautiously approached their keyboard. They type a few keys in, they click a button and they back away slightly because they don't want to disrupt it. They don't want it to fail. They don't want it to kick an arrow back at them. And that to me is just bad design. So from the ground up, we want you to feel safe. We want you to be able to find your information, put the right information in.
We want everybody in the system, in the company to be in the system without generating more costs for our customers. And we want to do that in a paperless environment. All of that is to drive more data into the application, because the more data we have, the better schedule I can automate for you, the better I can tell you what your margins should be. The better I can tell you what your utilization should be. And those answers are the ones that help you make more widgets with the same amount of equipment, the same number of hours, and the same number of people, which is what it should be doing. It shouldn't be this cave of dread where you throw a bunch of papers in. And once in a while you send your least favorite employee to go in there, pull a bunch of papers together, and co-lead together a report, right? That shouldn't be the experience, right?
Dylan Jackson: I use JobBOSS fairly frequently in my day job. And you are 100%, right. When I was first starting to look at your software. That's one thing that really stood out because like, I can't stand using JobBOSS because it just feels like I'm using a computer that I used when I grew up, you know? It's slow and clunky. And like you said, designed in 92, because it was like, it's, it's terrible. Whereas yours looks like a modern day application. So I definitely appreciate that.
Sunny Han: And it served a purpose before it helped a lot of businesses and it's still helping a lot of businesses. Our software, if we don't continue to innovate it, continue to push out the best newest features every single day, every week, every month, then ours will also be like JobBOSS. And at the core is our daily reminder of what software as a service means. Oftentimes software services, I'm going to collect money from you forever. But for us software as a service means, I want to provide to you the best software every month, the most stable, the coolest things. So there would be no reason you would ever want to use any other software. That's really our design ethos in our R & D. As we continue to grow is to not ever let any part of the application suffer out of laziness because people are fine with it. It's gotta be awesome everywhere over time.
Dylan Jackson: And that it applies to most of the manufacturing space. So it just seems like it hasn't caught up to Europeans yet. I mean, Autodesk made Fusion 360, it looked like AutoCAD. Nobody would buy it. So it's, you know, I think it's about time that somebody made something that looks nice in addition to being very usable. So I appreciate that you've talked a lot about all the features you implement. What does it cost for a shop to buy in or subscribe to Fulcrum?
Sunny Han: It really depends on your size and your complexity, but at the very low end, it's $500 a month. Setup fees are either nothing or maybe a couple thousand dollars to get started. It really depends on how much help you need migrating your data. Our philosophy is that we don't want to be a service company. I was a consultant for a long time. I don't want that job anymore and I don't want that to be a part of the deal shape. We have features that allow you to upload your own data, set everything up yourself, configure on permissions and roles and everything. We try to make that process, even though you only do it once ever, we want to make that process really good as well. And the whole goal is, is that if you look across the entire ERP space for manufacturing, I think something like 65% of all the revenue towards ERP software goes towards implementation consultants.
And so if we can eliminate that through automation or at least reduce it significantly, that just means more value faster for our customers. There are large customers of ours that are paying $10,000, $20,000, $30,000 a month. They're processing tons of dollars, like hundreds of millions of dollars of revenue a year. So there's a huge wide range. Our software really, there, there isn't very much that our software can't do right now. We have a big birth, but I would say that if you're a really small shop, we have a really great experience for those companies as well.
Dylan Jackson: So for those small shops, let's say just so we can stay kind of relatable to our audience. You know, a couple people shops, you know, maybe I don't know, five, 10 machines, max, you know, how long does it take to set up something like that?
Sunny Han: We have a customer that just started with us. He just started his business. He has three machines and he was up and running in a week, which we worked really hard to everything for these live in the system. Now running jobs through job tracking already. So, again, he's not moving from another system, right? He's a brand new business. So there's a lot of baggage that he doesn't have, but one week, four business days.
Dylan Jackson: Amazing. So you've talked about how it's a cloud-based system, how much of the data is stored locally and like, what happens if I have a network outage? I know that there's this constant push to cloud-based stuff. And that really worries a lot of people in manufacturing. And I think those are usually the first big questions they ask is, you know, how do I know that I have access to my data when I need it?
Sunny Han: That's a very valid point. And it's one where if you're, there are customers that we've talked to that are literally in the middle of nowhere, Montana, there's not even cell reception out there, let alone good internet. Some of them are still on ISDN or some sort of dial up. I think in those scenarios, frankly, we just wouldn't be a good fit. We want to store all the data for you on our cloud servers just eliminates maintenance for you. A lot of our customers go from paying an IT company.
If there are certain restrictions, we have customers where all of their drawings and big files are stored locally and our software can interface with it. We have to do a little bit of extra work to make sure we have the security permissions to do that. Typically speaking, you don't want your browser to be able to access your file system for security reasons. But the philosophical point is that internet access isn't getting less prevalent. It's getting more prevalent. Elon Musk is launching Starlink. You have your cell phone, you can tether to it. Our software is well optimized enough that you need a very, very slow trickle of internet to access it and work the system. So I don't like to sugar coat anything. Again, if you're somebody that doesn't have internet access, it is a non-starter, but even if your internet is weak or you have a cell phone backup or whatever it may be, or you just want to pull it up on your phone because we're, we're not a legacy architecture, you could use any browser device, you could use a tablet, you can use your phone, pull it up, log in.
You can, you can get your work done through there, which there are some advantages there to that.If you haven't worked like that, you probably don't have the context for, but there's some stuff that you gained too, along with the things that you do lose.
Dylan Jackson: That makes sense. So we had a few listener questions we posted on Instagram. I guess the first one was actually a great one to start with. So @tangentdarts asked when is the right time to start looking at any ERP? Is it, you know, people based, is it number of machines, a dollar value. You've had so much experience with this in the past, when do you think is the right time to start looking at one?
Sunny Han: I think the best answer is just whenever you start feeling like things are getting more chaotic than you want to handle, and that's the key, right? If you're a powder coder with customer supplied materials, you're not purchasing the very much you can handle, you know, you can have 20 employees and have only five customers, and it would be pretty easy to handle that, right. One of the best businesses that I've ever met was a machine shop in the Boston area. They made pins for medical devices. They had 70% margins, they had three guys operating 12 machines, huge revenue, but, didn't need any software. And they just made a killing. Like if I could engineer that for myself, I would probably not be in the software business.
I just retire off of that. Not everybody can get those relationships and do that type of work. So really the answer is much of a cop-out answer as it is, is it's when you start feeling like it's too chaotic to manage, or you feel like there's something you're missing, or you're, you're, you're lacking control for most people that are manufacturing control. This podcast is called Within Tolerance, like accuracy control management. These are things that are second nature. So whenever that starts to feel like you don't have that control or power, you think that there is a more efficient way to operate and you just don't have the right tooling for it. That is the right time to start looking, even if it's not the right time to buy it is the right time to start looking.
Dylan Jackson: All right. And so I guess a followup question you've, you've touched on machine learning and automation and all of that. I feel like machine learning is kind of a buzzword. I'm not sure that everybody really understands what it actually is. What does Fulcrum do to make my job easier? Like what you talked about it back feeding into your quoting system. As a small shop, let's say we're a two person shop. What tasks is it going to take off my plate that I can just dedicate to the system?
Sunny Han: So at a very small level and just explain the technology behind what everybody kind of bunches together, and as artificial intelligence, most of these concepts have been around since the 1970s. These are mathematical conclusions to analyzing data, to get good results that humans can't really find the answers to very quickly. We just simply haven't had the computing power or the engineering technology to put it all together and make it work. Specifically for manufacturing, machine learning is basically if you had a infinite number of super smart math nerds, every piece of data and correlating every piece of data with every other piece of data in your whole business, even a two person shop has hundreds of billions of combinations of data. Machine learning is about doing all those calculations in the right way to find the right relationships and say, “Hey, if you do it this way, you're going to make more money.”
If that's what you're optimizing for, if you do it that way, you're going to make less, and maybe we should steer you towards this direction versus that direction machine learning can't actually solve job shop scheduling. It has the spooky way of being able to tell you what the make span is and make span is basically how long under optimal circumstances is going to take you to do all the work on your backlog machine. Learning can tell you that, but it can't tell you the order to which to do it, and when to do it and what the schedule together, et cetera, et cetera. There's other algorithms and things that can be used to do that. But for a two, three person shop, the real answer is at that stage, you can solve your scheduling problems in your head on a scratch piece of paper.
I operated a small machine shop for a few months, and that's what I did. I mean, it's embarrassing, but I just had like a notebook and I would look at all the jobs they would, I would just, I printed them out and put them on file folders and I would sketch them out and look at their operation time. And my scheduling was done in 10 minutes. Right? So there's nothing that we can do at that size. That's better than what you can do by hand. Now, when you have, you know, 30 work centers, 15 work centers, 10 employees, a hundred jobs, 200 jobs. Now we can start to deliver answers that are really nice, that are 10%, 20% more efficient than a human could get to. But even at the low level, just imagine that you put data in
Once you entered the quote, once you convert it, you don't have to type anything else to add some information about the sales order. It tells you, this is what you have to purchase. You click a button, the purchase orders are done. Once it's purchased, we have a screen that says, here's everything you were supposed to be receiving. You receive it, take a picture, all the data's in there. When you start the job, all that information is in there. The picture of the material, as it arrived is in there, you enter your quantities done and all your paperwork is done. When you ship it off, you press a button and automatically integrates with the shipping companies. You produce shipping label, pack list, bill of lading, all that just comes out. You put it in your packaging, you ship it off, you press one button. Your invoice is done.
So having the information in one place and just trusting that it'll be there when you need. It just allows us to automate things that otherwise would have been, open up QuickBooks, create an invoice, find the right customer, add the items, massage it a little bit, turn it into a PDF, send it off, you know, look at all of the things, export it into a spreadsheet, take a look at what you need to purchase. Make a few purchase orders, you know, kind of juggle it around and figure out which vendors you want to buy from, and then make a purchase order entered into QuickBooks, receive it, and then receive the bill and put it in those tedious things seem, kind of trivial, but they add up and really for the smaller shops, the outcome is you could stay a smaller shop and not higher than offers person, not hire that next person and have 40 50% more revenue before you kind of have to scale up to that, that stage. Whereas traditionally, you have to hire somebody, some help that eats away at your margins. It eats away at your break. You just have this nice, comfortable zone where you can make a little bit more money and last a little bit longer before making it to the next level.
Dylan Jackson: And then we got another question: what are your tips and advice on proper ERP project implementations? And then he says, many ERP projects are doomed to failure or end up as a hodgepodge solution, similar to what it was going to replace.
Sunny Han: That's a great question. And it's a problem. That's getting worse because in the sixties and seventies and eighties and nineties in our society, we did a lot more things that were joint. We got into a car and went to go see a movie. And we had to agree on the movie that we saw. Now we can stream whatever we want. We can play whatever games we want. We can play online on Xbox. We can play on our phone. We can watch videos on cooking on YouTube. We can do whatever we want whenever we want. And the core goal of implementing a systematic software, like an ERP is to get to a point in time where we all agree. This is what we're doing. And the bigger the company, the more risk there is that politics will ruin it.
And with anything, anything that touches that much of your business, there's always going to be people that lose. And there's always going to be people that win in the short term. Everybody can win in the longterm. And after a little bit of adjustment, this isn't a zero s system. It isn't because I have something easier means your life is worse. , but getting that commitment upfront is oftentimes really difficult. We can do a lot of things by giving somebody, everybody giving everybody in the business, something to look forward to. That's what our job is, but the job of our customer is to commit to it and say, we're going to get all the fundamental data in there to begin with. I think there's three big things that drive failure in ERP implementations. The first is just a sense of lack of confidence. Sometimes you just have to have somebody say, look, I know you're a third of the way through changing all of your, , you know, all of your items that were in keystroke, which is really old inventory system and putting it in a, in a nested format so we can roll it all up.
I know this is tedious. You're a third of the way through keep going. It's only another day or two worth of work. It'll be worth it. And oftentimes that in and of itself is really valuable because imagine if you just,, start off with your top 10% of items in inventory, and as I get orders and other items, I'll add them in at that point in time. Unfortunately, that never happens. And it never works out that way. Right? You all know that that's how these hodgepodge solutions, like you said, happen, is that we have a hodgepodge commitment to getting the things started as quickly in the right way. I think the second is, is that there is a lack of kind of understanding on how the whole business is connected before. If you just had spreadsheets, you just had your own little bubble to do whatever you wanted to changing something didn't affect me.
It didn't affect you. Now, as we start using a system that is connected, there's a lot of value and benefit in there, but there's also a lot of potential controversy. It might now expose the fact that I do all my work at the end of the day. It might be that I do something wrong all the time. People can now see everything and, and there's a resistance to it. So I would say number two is just making sure that everybody understands that no, one's going to get in trouble. This isn't about firing people. This isn't about making people look bad. It's about making the company operate better, getting people to really believe that and feel safe is really important to having a good implementation. And then lastly, making sure that you are the one that's deciding what processes in your business are important to keep, and which ones are bad, the consultant that's implementing it, the software sales person that's selling it.
They can't possibly have no matter how much they try your full interest at heart. They just can't understand your business. We train our people to ask a billion questions to the point where people get frustrated and annoyed, but our goal is to really understand as much as possible even then we still miss stuff. Right? And so I think it's important to let us be the experts in software, but also to recognize that you're the expert on your own business. And that doesn't mean that you don't have to change. It doesn't mean that your bad habits are ones that need to be kept. But it does mean that if you feel strongly about something, even if you end up being wrong, it's important to stand up for those things you believe in because there's some pain, there's some customer that didn't pay you for three months.
There's some rejected parts, pain that is behind all those stories, right? Knowing that might not lead to the answer that you want, but might lead to a bunch of other things that this person that's helping you implement. The software can just act as if they are you, so communicating those details no matter how embarrassing they are, even if it's like, you know what? Yeah, I actually stick all the checks that I get in a drawer. And when I need money, I go to the bank and deposit them. You know what that has been there's and you probably shouldn't do that, but tell us, and at least we'll know, right? At least we'll know that that's the cadence in which checks come in, not to expect that you deposit the same day. So I think that last piece is the most important, really open and honest communication.
I'm not here to judge you. None of my team is there to judge you. I hope there's nobody out there really judging people for doing things one way or the other. Our goal is to help you make more money as quickly as possible. And to do that, we need to know who you are. And if that's, if it's there, if, if I'm honest about whether I can do it or not, and you're honest about whether you can do it or not, we can, we can do great things together. So, long-winded answer. But I think that those are probably the top three reasons.
Dylan Jackson: Awesome. Another question we had from Machinist Brett was it seems like almost every cloud based software has some lag to it, you know, click this button and it lags, click that button and you have to wait, how does your software minimize that?
Sunny Han: Yeah, but better engineering, just like certain parts are made better by certain people, certain pieces of software are engineered better by, by other people as well. We have really good talent. Our engineers are from Bloomberg or from NASA or top one 10th of 1% engineers or designers are really great. We're looking at things, we're inspecting them. We want them to be shiny and pretty. We want them to be uniform. We want everything to work well, and we want our buttons to have under 200 to 300 millisecond response. That is the goal that Google has. And even still, if you go search something on Google, it'll tell you it searched 15 trillion, something in like a 10th of a second or something like that.
Right. Then if Google can do that with 10 trillion web pages, your ERP should be able to do that with 14 items in your database, right? Like that isn't, that isn't a mathematical constraint. It's just bad engineering. And that doesn't mean that we don't have that engineering in places. but we have it in very few places, not because of any magic, secret sauce, it's just attention to detail. And dogfooding which in the software world means using your own software. And this sounds silly, but we get together once a year. And last year we started a tradition where we act like a bunch of manufacturers that are working with each other. We make paper boxes and our paper boxes look terrible. They're there. They're not made very well at all. But it allows us this kind of concept of all right, this operation has to happen first and here's some outsource operations.
So we role play as if we were working for a powder coater for a five-axis CNC machine shop and for, you know, a laser cutting and brake press operator. And we have to put everything together. We do all that, so that when you push a button, there's isn’t a lag. And the only reason why in any software that would exist is that either the engineers never pushed the button, or they never pushed the button with real data. So it seems super fast because there's no data in the system, or it's not valuable enough for them to fix it. So there are companies who have been purchased by private equity funds that literally there's one software developer left on staff, managing the bug reports for 5,000 customers. In that case, you gotta have a little empathy for that guy.
He's not catching up with button click lag. So, that's probably out of the question, but there are other companies that, that kind of don't care, right. I think there was a study done in 2019, or sorry, 2016 that showed that something like 70% of all manufacturers extremely dislike their ERP system. And yet those same populations still kept that system for an average of 7.1 years. So that says something right as a potentially gross software company. I think I can subject you guys to quite a bit of pain if I want it to, and you'll still stay. So there might be some disincentive to do the right thing but I think fundamentally it really is not to do with the internet or to do with how applications work is just an engineering question.
Dylan Jackson: Okay. You were talking earlier about how you guys are pushing, you know, new technologies and all that. What does the update cycle look like on Fulcrum? How do you guys bug test it? You know, do you have a beta test group or like, is it all automated? How does that all work with you guys?
Sunny Han: We have everybody sharing the same code base, but they're all separate. Some of our company, our customers are ITAR compliant companies. So they have to keep their files separate. And when we push out an update, we do a rolling update. So we have a demo instance. And then we have a test instance, we deploy our code there, we do a little bit of testing. Then we roll it out to people as it affects them. So if we're just pushing out a new shipping interface, that's gonna change everything, we want to make sure that the first person to test it, the first customer just gets it. They look at the interface, it freak out for a second. They're like, wow. I actually just understand how to use this new interface, even though it looks completely different.
If that isn't the case, we're going to roll it back, redesign it until that is the case. That's how we ensure that the interfaces are intuitive. Even if we go from a grid to a much more decorated grid, if you don't immediately understand the new thing, even though we were all used to the old thing, we haven't designed a new thing well enough yet. So as we continue to go, we monitor that, we look at people, are they even clicking that button anymore? are they doing it outside of the systems were really paranoid, maybe a little creepy. We're analyzing all those data, but really we're trying to make sure that people are using it and it's bug free and we're rolling it out in sync. So unlike some other software, our software, doesn't kind of deploy to everybody simultaneously. We can select different clusters we deploy to and roll it out over time.
Dylan Jackson: And then how, I guess, how often nothing to hold you to. But how often do you usually see upgrades happening?
Sunny Han: If Machinist Brent, he calls me and says he clicked a button and it took a second that that changed. We pushed out by the end of the week immediately. So there's, there's different buckets of things. If it's a bug performance improvement, Polish, if it doesn't look right on a mobile device or something. We're just going to push that out. No, wait, if it's a new feature, we push those out. We, we, we try to have new features out once every month. And if it's a revision to a feature, a Polish, that's typically every six weeks or something like that, usually on that cadence.
There's a lot to build. There's a ton more to do. We're going to, we're going to automatically suck data from your machines. We're going to give your customers and your vendors, automatic portals to be able to integrate with their QuickBooks so they can submit orders and keep all that synchronized. We're going to have chat inside the applications that you can reference drawings and leave notes and talk to people. There's a, there's a bunch of really cool stuff to build stuff.
Dylan Jackson: That's awesome. So the last listener question we had was from Brandon over at a Machine Shop Outlaw, and he said, is this, or any other software, good replacement for parts and vendors? And I'm not sure if you're familiar with parts and vendors. I know I wasn't had to kind of look it up.
Sunny Han: If it's the one by Trilogy Software that, that I'm thinking of, yeah, that I haven't heard that for a long time. So it's an, it's an old system that was, that was used by a lot of like prototyping labs by scientists, like scientific institutions. I think for me, labs used it and a bunch of universities did, but yeah, I mean, our, our functionality is way wider than that, but we handle revisioning, you know, multiple vendors per each different parts, different BOMs that are nested on each other that are sub assemblies that exists as inventory items, as well as components within items. And you can kind of expand the entire BOM..
Dylan Jackson: I had a question looking at your website. So I'm a self-proclaimed quality nerd. I started at my day job as a quality personnel and I still, you know, program our CMMS and, and deal with a lot of the quality issues. Can we talk about your NCR system and your KAPA system and what kind of data you can pull from it and all of that?
Sunny Han: We want to automate as much of the data collection. So if I'm spawning off a nonconformance, I should know that you were job tracking this job for this customer, with this material. My system knows where you purchase from who you purchase from. So a lot of the kind of formality of tracking all that information just becomes very automated. We just pull all that information. You can edit it if you want to. But for me, my philosophical take on quality is, is, is pattern recognition, it's monitoring and it's trend management. So you want to know if there's any hotspots, right? You want to know if it's always this material, always this vendor, always this machine, always this combination of those factors, that's a lot of where machine learning algorithms can come in is to analyze a bunch of those sub correlations to say, actually every Monday afternoon, sunny working on, anodized alin on the five axis.
CNC always messes it up. Oh man, how did you know that he's sneaking back home, drink a half case of beer every Monday? Well, the machine told me that, right? So our core philosophy is we want to match with industry standards. We want to be compliant with them, but really the more data you can collect, the easier you can collect it. The easier I can make it for you to put in what those tolerances should be to give the operator live feedback that says “Hey, this last part is getting dangerous to close tolerance. It's drifting.” I can make a change to the way that the setup works or something like that. That in process tracking is what it's typically called statistical process tracking, but really it's giving live feedback to the operator so that we prevent mistakes from happening before they happen.
And then over time, we're tracking all that information and running reports on that. I'm on Reddit all the time. There is a subreddit called oddly satisfying. And I always look for like the wire EDM, the GIFs where you can see the thing lining up. So it looks like a brick of metal, but when you pull it up is actually five different parts. I love that. I love seeing things that are really uniform and well-machined. I think there's a sexiness to that that draws people into manufacturing, machining, right? Like, I think everybody hopefully adheres to that and, and really gravitates towards it. I, I don't, I don't think that I don't think that people like making dark carts, it's the times where they don't know, or they can't tell, or they could have predicted it, but they didn't.
Those are the places that we want to inject technology to help people. I think fundamentally these systems are based off of processes and philosophies that were invented to help Japan in the fifties and overcome their quality issues that reported back here, bounce back and forth. There's a lot of baggage and, and hand-waving, and it sometimes seems like a wizard with a pointy hat reading a spell out of a book. But I think fundamentally we can remove as much friction as possible and get to the meat and potatoes of the data. Our job is to track that data immediately, automatically, and associated with your nonconformance reports and then help you generate the correct corrective actions and preventive actions wherever possible. But fundamentally the real quality changes that happen are providing the operator of the data live in the job tracker.
Dylan Jackson: Gotcha. So in your example if your system starts seeing these patterns, you know, let's say you screw up every Monday for a month at this time, is it going to ping you or are you going to have to run reports?
Sunny Han: It, it doesn't, it doesn't ping you yet, but we are adding in text messaging support soon, so that, that might actually come. But, right now the primary methodology is we have a default heat map report that allows you to analyze any factor against any factor in to us and to the people that have seen it. It's just the easiest way to scan across and your eye can pick up anything that looks like an aberration. You can see if it was that day, everybody was bad. Or if that really is a kind of a sore spot for that person. That's how we let people analyze it. And we default the report to the most variant combination of variables. And then you can kind of select how we want to slice and dice it after that.
Dylan Jackson: Okay. Awesome. And then you briefly touched on in-process inspection. Can you put in-process data into your system yet? Or is that a future thing or?
Sunny Han: Yeah, setting up checklists and tolerances and different checkpoints and pieces of data and units of measure in there. Those are all things that you put right into our job tracker, and it stores it. And if you need to print it out with the pack list, you indicate that on, on the item master, and it just prints all that stuff out or emails. We have a company that we EDI automatically into their customer's ERP system. So, yeah, when you store that data and we pipe it to the places that it needs to be,
Dylan Jackson: And you can run CPKs and all of that from the data too?
Sunny Han: Yep. And your data is yours, so you don't need to buy any fancy packages. We can give it to you in Google sheets, in Excel, you can play with it however you want to. All your data is yours to play with and report on however you want to.
Dylan Jackson: Wow. That's super impressive. That's like a million different software packages all rolled into one. An area I've been looking into quite a bit. So I'm impressed that you can do all of that through the quality. That's fantastic.
Sunny Han: It's been tiring. I need to sleep at some point in time, but there's still more stuff to build.
Dylan Jackson: And so for our listeners, who are ITAR compliant, you guys can do an ITAR compliant install and all that as well.
Sunny Han: Yeah. We can deploy on Amazon gov cloud and Azure gov since Microsoft won the department of defense contract, everybody seems to be on Azure gov, but yeah, we can, we can deploy on a government secure.
Dylan Jackson: That's awesome. That really opens up the market for you guys. I know that's the hindrance to a lot of cloud-based software right now.
Sunny Han: Yeah. And you know, that should change. I don't think that's unique to us. I think Amazon, Google, Microsoft all have a lot of vested interests in making that as easy as possible for companies like us. So I would expect that in the next 10 years, it'll be easy for everybody to be ITAR compliant. So just from a global perspective, if you're an ITAR shop, I would not have any fear about that being an issue for very much longer, whether it's with AutoCAD or you know, or any other software package.
Dylan Jackson: Awesome. Well, I also wanted to plug that you guys have a great COVID 19 relief page on your website. So anybody who's still looking for more information, I know Tyler sent it to me and you guys kind of lay it out a lot more clearly than a lot of websites. So anybody who's looking for more information. I mean, I think a lot of the loans are closed and stuff, but if you're still looking for PPP or economic disaster loan information, they have a great little page on it, on their website.
Sunny Han: Iit was written kind of casually. It was like three in the morning. I was getting frustrated, reading it myself and I just tore it all apart. And we wrote it in better words, I think. And there actually still is I think 150 billion more dollars in the second re-up of the PPP loans. So for anybody that has disruption to the revenue stream, I would certainly encourage you to continue to look into the documentation. And I would certainly prioritize any local banks. We have seen the most success even with local credit unions, with our customers where they didn't have any previous banking relationship. Most local credit unions and community banks have been really, really generous about allowing people to have these loans processed through them without being a previous customer. So I think that if there is any type of inside tip the best one would be to seek out community banks and credit unions.
Dylan Jackson: Yeah, I've heard the same thing. I know there's been a few people who have posted on Instagram and stuff about how, you know, they've been a customer of B of A, for 20 years, for example. And they were denied because they didn't have a credit card with them or something like that. And so it seems like the local credit unions really stepped up to kind of close that gap of people being able to get it.
Sunny Han: And you know, this is a terrible environment that we're in, but I'm just super excited. I think if there is one big positive for manufacturing is that the supply chain, cracks and shards are completely exposed. There's a lot of attention on it. People know these buyers at big Fortune 500 manufacturers who are told never to shop around for new OEM manufacturers. They're they're shopping because they have to, and they know that there's risk there. So I think this is a great time to have knowledge and expertise in machining and manufacturing as good of time as any in the last 15 years.
Dylan Jackson: Well, I think that wraps up our questions. We always give our guests some time to plug all your socials or plug your website or your business and anything that's coming out soon or new or anything like that.
Sunny Han: My email is Sunny. Like the weather sunny@fulcrumpro.com our Twitter is twitter.com/fulcrumhq. You can find us in any of those channels. Our website is fulcrumpro.com. I just like to say that if there's anybody that has any questions about manufacturing, about software, about technology, if there's anything that is bothering you, that you feel uncertain about, I want to make sure that I'm contributing to this effort that you guys are doing. To me it’s about making information as accessible as possible. So if there's something that you might not want to ask publicly, or you think that I might be able to help with we as a whole company are committed to giving as much value, benefit, knowledge assistance to this marketplace as possible. So our policy is to just email me. I'll try to respond within hours if not within the same day and anybody in our company will do so as well.
Dylan Jackson: Well, thank you so much for coming on. I know that ERP is kind of like you said, it's a dark cave that a lot of people don't want to go into. So it's great to have somebody on with your breadth of experience that can kind of shed the light in there and then show that there are companies looking forward and moving forward with that kind of stuff. So thank you for your time and thanks for coming on. We really appreciate it.
// This article was update 4/04/23 with the help of Ai.
