SiliconExpert Podcast Episode 4 with Admir Kovacevic of Bay-Soft - Transcription

Host: Eric Singer

Producer/Director: George Karalias

[00:00:00] Admir: I was exactly six months in the development department for an OEM and until the owner came to me and said, I have to tell you, usually we don't hire and fire, but I have to tell you, you are one of the worst engineers I have ever seen before. That's quite a slap in your face. I have to tell you.

[00:00:19] Eric: And while Admir had aspirations of being an actual rocket scientist, his talents lay elsewhere, his benevolent boss, who just minutes before had fired him connects him to a path that would lead him here.

[00:00:32] Admir: My name is Admir, and I'm head of sales and marketing in Bay-Soft. And recently also a partner we offer highly integrated, Bay 2 PDM and Quotes system for OEM and EMS.

[00:00:45] Eric: Welcome to the Intelligent Engine, a podcast that lives in the heart of the electronics industry brought to you by SiliconExpert.

[00:00:53] SiliconExpert is all about data-driven decisions. With a human driven experience. We mitigate risk and manage compliance from design, through sustainment, the knowledge experience, and thought leadership of the team partners and those we interact with every day expose unique aspects of the electronics industry and the product life cycles that live within it. These are the stories that fuel the Intelligent Engine.

[00:01:20] Today's spotlight is on Bay-Soft a company based in Germany with the vision and passion over the last 15 years to automate and digitize with their software Bay 2, the PDM and SCM process. With its headquarters in Germany Bay-Soft offers the highly integrative Bay 2 PDM and quote system for OEM and EMS, which is networked with the various systems such as ERP, ECAD, and external databases, information on price, availability, compliance as well as form fit and function specifications are shared from external databases. We have Admir with us today, head of sales and marketing at Bay-Soft. We'll talk about digitization and IOT in our industry. Admir's background is quite wide. Over 20 years ago, he started as a semiconductor engineer and developed industrial applications for a European OEM.

[00:02:12] His journey brought him from the technical to the commercial world, where he was responsible for the sourcing department within a mid-size EMS. The next career steps he made were within the sales department for a distributor before he recently became a partner at Bay-Soft. Admir, thanks for joining us.

[00:02:29] Admir: I'll would like to thank you for having me here. It's a pleasure.

[00:02:33] Eric: So quite a start to your career, but your boss saw something in your talents. So it's really more of a mentoring redirection, right.

[00:02:41] Admir: Instead of developing and instead of playing around with all the tools and gave me something really cool and fantastic, you keep asking price, so who's going to pay that. Why should we have this? Can we have this? So is that you're honestly speaking, you should leave everything that is technically related? Because you're more commercial related. So if you want, I can suggest you to a friend of mine. He has a company EMS, and he's looking for somebody for the sourcing.

[00:03:06] So that is how I had a short period of my life being an engineer. And change into the sourcing area. Yeah. And from that one, I started like the typical way I was a regular buyer, went through different departments, responsibilities, commodities. And after four years I actually realized that the sourcing is very nice, but I talk too much.

[00:03:30] Eric: So sales!

[00:03:31] Admir: Sales! Exactly. Then I started for a distributor as a sales account manager. Yeah. My last role was sales director. And during my whole career, I really speaking. I was always let's put it this way. I am. I'm a real sales person, a real salesperson means, honestly speaking. They are the laziest people you can meet at least in my world because seriously, we want to efficiently make money to sell whatever, but we don't want to spend 24 hours. Just, these are the developers. These are the engineers, they are the love to code and they love to, to develop. Sales is more towards let's, have fun, play golf, do something and make money out of that. So in the past 20 years I always tried to find ways how we can optimize what we do, how we can use the tools which are available to make us better in sense of how can we work fast or how we can reduce the error ratio? How can we communicate better without typing tons of emails? So try to find a way through my career, I always like everyone. I know we started with Excel access databases and whatsoever. And, uh, so that is how I came to Bay-Soft because a friend of mine, he said, honestly, speaking, you have to watch this company.

[00:04:53] It's a phenomenal, good tool. Like they address exactly what the industry is looking for and they just have one problem. And the problem is the sales. Because at that time we didn't have a lot of customers. We had a niche product for niche industry, if less amount of customers. So that's how I came to Bay-Soft became a partner and said, you know, guys, I definitely cannot code and I definitely cannot help you in developing whatsoever. I tried. I proved. I missed. But, I can sell. I can try to sell. And it's few months is August. I joined and I have to say due to the effect of the Corona and the COVID situation, many companies wants to go towards the digitalization. So it's right now, it's a good time also to talk about that.

[00:05:40] And I would say in the past six months, we gain more customers than the past five, six years. So that shows that it doesn't require a sales person, it just requires somebody who talks a lot and presents.

[00:05:53] Eric: I love the comparison that you draw between the lazy salesperson, who wants to make their process as efficient as possible so that they can make more money and expend the least amount of work that applies, it seems to me perfectly to what Bay-Soft actually is providing. You want to make the process more efficient. You want to make it less work for the people who are using it. You found the perfect spot.

[00:06:22] Admir: The challenges, and really that's the interesting part. If you see our industry today, decision-makers even on the operative level, people are more responsible for, let's say a higher amount of tasks and you have to document, you need to keep the documentation properly.

[00:06:35] So if you ask people and that is what I ask, always ask the question. When I go into a meeting, it's like, tell me, how much do you act and how much you react. So how much you act meaning that you have a technical background? You're an engineer. So how much of your time you use your brain to come up with something to decide, to measure to whatever.

[00:07:00] And how much of your time do you need to do fulfillment? Like you fulfill data, right.

[00:07:07] Eric: Responding to requests.

[00:07:09] Admir: Exactly. Exactly. So when you started the conversation, should everyone will say no, no, no, no. 99% is just brilliant things all day. Exactly.

[00:07:20] Exactly. But when you dig deeper into it, then you realize that no, they are heavily depending to collect data.

[00:07:28] I need to check if the data is valid, I need to link the data to my old system somewhere. So that's my vision and that's the vision of Bay-Soft as well. So that's why we matched each other to say, no, we need to come up with a product which allows to people to decide. And let the engine work.

[00:07:49] Eric: Yeah. It really turns the process around to go from a reactive process to a methodology where you're driving the process from the ideation forward. That's a pretty huge shift. Is that something that we see more widely across the industry these days? Because obviously the old way of doing things, particularly in very large companies has to be very entrenched.

[00:08:12] The bigger the company is the bigger, more complicated their systems are and the harder it is for them to change. Is Bay-Softs, relatively small size, uh, has that enabled you to react faster and to be able to think more openly about, Hey, here's how we can change this entire process.

[00:08:29] Admir: Honestly speaking, I thought at the beginning that because we are so small, we can change few things fast.

[00:08:36] It's not about that. And I realized that it's not, you know, you have two ways of softwares today. One way of the software is that let's say, want to fulfill the requirement of the customer. And you don't care about how you transmit the data. Where is the data who is responsible for the data who proves the data and these kinds of things?

[00:08:55] Right? The other way is what we have is we take an architecture. Our architecture is from Microsoft dynamics space where we say, okay, first of all, we need to have an architecture where it does matter to which company I go tomorrow. They will not complain because do we control their data? No, we don't. Um, and we implement our software in your server, in your system, right?

[00:09:19] It's my software, but it's your data and you control the data and the architecture and the let's say this coding beside how to connect each element makes those things so complicated. So what makes us different is it's very simple. 15 years ago, even longer, the crowd from started this thought process and coding process.

[00:09:41] So what we coming up today, we started with developer two years ago. Right? We see here and there, competitors who let's say attack the same case, but it's not really, let's say down the road, it's not like, really, you go very deep and say, okay, have you considered everything possible? And when I say consider everything possible, I want to make it very simple.

[00:10:04] I cannot convince a customer. I cannot convince a user to use my software. If he cannot rely back on his chair and say, Hey, I don't need to control. Everything is done. So it's like when you are in this school and you calculate two plus two is four. So if you want to control your four minus two should be two.

[00:10:24] So we take this decision. We take this time away from the user and say: Lie back. Everything will be automatically and digitalized, and then you will get also the controlling function. So you will get the end result controlled. It's like a number, right? Yeah.

[00:10:41] Eric: You don't have to worry about two plus two. We handle that.

[00:10:45] You just ask your question.

[00:10:47] Admir: That is the difference. So that the one thing is they automate my process. That's the easy part.

[00:10:52] Eric: Yeah. Yeah. The semiconductor industry certainly is a cornerstone of the high tech industry in general. Does IOT exist in our industry besides on the manufacturing side?

[00:11:06] Admir: First of all, in general, let's say the new system, current ERP systems, all of them, I would say are IOT capable and to many of their users, many of the clients in our industry, they are maximizing it already.

[00:11:19] So for example, an ERP system is mainly for financial logistic and for the production relevant, the operating part of sourcing, like creating a purchase order, a variety of different logistics solutions, documentation, all of these things. They're very advanced in, in the process flow of current ERP systems. And it's very efficient.

[00:11:42] Yeah. The things that we do thousands of times a day across thousands of different companies, we're pretty good at that piece of it. Exactly. There is no difference between, I dunno, producing of high-tech product versus milk, including milk package. They have both the same ERP system. And the process in the ERP system is most, probably the same.

[00:12:04] Eric: Pretty basic stuff, no matter what you're building.

[00:12:07] Admir: Correct. So if you see it from the ERP side, yes, it exists the extremely high amount of variable product data we have now in the semiconductor industry includes all the product changes we get affected every day. The concept of change market conditions, price, lead time, different venders makes the whole thing complicated.

[00:12:28] So the. Yes, we have IOT in our industry, but only for financial logistic production areas. But when it comes up for the development department, when it comes up for the sourcing department, definitely not. So if you ask a buyer today, how his daily work has changed today versus 10 years ago, many of them will say the past, I had less amount of emails, less amount of part numbers, less amount of complications and headaches. But the tools are quite similar, which they are using internally, Externally, it's an absolutely different world. Externally, like we also, if you see in SiliconExpert, there are platforms which has drastically change and help in our working life, but they are not connected to our system. So IOT, in my opinion, in our industry related to development and sourcing department, does not exist.

[00:13:22] Eric: Yeah, that seems like a big hole to fill.

[00:13:25] We talked about an example of making a high tech product versus let's say packaging milk. So if you're a dairy and you're packaging milk, you probably don't have to update your products. Very often, milk comes out of the cow. You put it in a carton. Maybe you change the look of the carton. Maybe you even change to a different kind of carton, but there's not much else to it that's going to change the things like form factor and all that. Can you talk more about what makes it so complicated in our industry specifically?

[00:13:57] Admir: Okay. Let's start from the technical part and the commercial part. Let's talk about the technical part first. So let's say now I try to address my engineer friends.

[00:14:06] We get confronted with several different areas, which we have to fulfill. So one area is as example, what comes from each country, from each governments, what they ask to fulfill. So, as an example, you know, we have to fulfill the Dodd Frank Act, the dual use, they reach, the RoHS, Reach 2, and whatsoever you have so many, um, datas, which you have to collect and to update your system, just to meet.

[00:14:36] Um, requirement's

[00:14:38] Eric: regulatory. Exactly. Right. Exactly.

[00:14:41] Admir: Then the second thing is which comes from the manufacturers in the past, let's say decade, we have seen quite a lot of acquisitions in our industry. So each acquisition brings few changes with it. So either the manufacturer name change or the product name change, or the product goes obsolete.

[00:14:59] Now there's a few changes there, which we cannot influence where you cannot even as a developer, you cannot even know what's going on because that's confidencial. So suddenly you get confronted and said, okay, now you do something right. And beside of all of these two things, there is the third thing. And that's what you have said.

[00:15:18] Say, you have milk and you will have 10 cows, a hundred cows thousand cows, right? If you have few hundred thousand core part numbers, and these part numbers have several slight changes in the part numberings or suffix prefix. The one is related for the package for the temperature grade. So these kinds of huge amount of data as an engineer, you cannot know it.

[00:15:42] You need to have a database. I remember when I was an engineer, honestly speaking, we had a whole room with literally 200 square meters of datasheets. So when we had to collect the data sheet, it took us like one hour to, to get these things like today. Faster, but today you have also much, much more part numbers which you have to cover.

[00:16:03] Eric: Yeah. And, and when a part number changes, who knows if that's going to affect anything that will change what you're doing, maybe it's totally irrelevant. Maybe it makes the part useless for your product.

[00:16:15] Admir: Absolutely. The keyword is workflow. So this thing is that one thing is what happens. And the other thing is who has to do what?

[00:16:23] So a clear workflow, but that is then that the next part of how we solve it, like, or how should a software solve it in, towards an IOT. And on the commercial side. That's a different ball game because if you see the extreme fluctuation in our industry now, or now the change of you can read it everywhere the newspaper that there is a shortage in the semiconductor industry. Tomorrow, it will turn around.

[00:16:48] So you have to constantly change. That is that a static thinking and a static workflow doesn't exist in our industry. And that makes it first of all, so beautiful. That's why we are here where we are. And we love it. But on the other side, that is also the challenge we have to find how to make it easier. As it is today.

[00:17:07] Eric: Yeah. There's enough challenges. Let's talk a little bit about the research side. We've talked about regulatory compliance. We've talked about the commercial side. What are the challenges to implementation on the research side?

[00:17:21] Admir: Where to collected data now that's the thing. So I'll give you the perfect example.

[00:17:25] Let's say I've worked with 10 different customers in one room, and I would ask them guys, how do you collect your data? Where do you store it? How do your people have access to this data? Most probably eight of 10 will have a different idea. We want to talk about one process and how can they optimize and make it better?

[00:17:41] Eric: Yeah. Yeah. Stop wasting time.

[00:17:44] Admir: Exactly. So what is the biggest challenge? When it comes up to collect data? ERP system is limited in the capability of collecting data. So this ERP manufacturer did not think about Reach 2 a few years ago when they came up, uh, out as example, right. Or maybe pick and place data or these information.

[00:18:04] So you need to go to the ERP manufacturer and ask them to add these fields in the database. That is the easiest part. Cost wise will be a different thought, but that is easiest. We talk only about products, which we are using for our production, but how does it look like for EMS? Like a typical contract manufacturer.

[00:18:27] He gets a requirement from 10 customers and maybe one or two orders he will get. So that means that from 10 customers he gets the bill of material, what to do with this bill of material. You cannot put it in your ERP system because you blow up the ERP system. It doesn't make sense. So mainly all of them, what they do is they have like Excel sheets and do upload and download.

[00:18:51] And here comes really the sad part. You have, let's say mid-sized and big contract manufacturers. They have several departments in several areas. And they request the same part number and they ask the same data for the same part number or collect the data for the same part number. But they don't see it because each side or each user has his own Excel file and they don't import it into the ERP system because there is no reason because we are still in a calculation process. Right? So what we do with our software is that say, no, no guys, we put you all together into our software Bay 2. Right? This software, Bay 2, is the bridge between your ERP system and the data provider outside, for example, SiliconExpert.

[00:19:39] So if your friend in Mexico collected the data for this part, your colleague in Germany will see it as well.

[00:19:48] Eric: And he doesn't have to be emailed an outdated spreadsheet that may not make any sense to him. And yet we've eliminated 10,000 individual Excel sheets across the world.

[00:19:59] Admir: And the most beautiful thing is today we calculate the bill of material.

[00:20:04] We quote to our customer to the OEM, right. And suddenly there is a product change notification coming from the manufacturer. Right. So ideal case in my view, What do we fulfill with our software is that the PCN goes into our system. It goes directly to the buyer or to the salesperson who quoted to his end customer.

[00:20:25] So even though you didn't get the order from your customer, you can already drop an email to a customer say, Hey, by the way, I quoted you yesterday, your BOM actually today, a PCN came up, right? So maybe you should not consider this product into your application and this service pitch makes it today different.

[00:20:47] Eric: Yeah. That's a real paradigm shift. Getting back to what we were first talking about acting versus reacting.

[00:20:54] Admir: What I said at the beginning, you have two different approaches, how it can develop something, you can just solve something that requirements. So the customer wants to automate his quotation or calculation process.

[00:21:04] That's one thing. Our approach is a little bit goes more deeper because if data is available, then I want to have this data specifically nailed to the responsible person. Doesn't matter if it's the buyer developer, sales, and if you have these arguments, then you're, you represent a completely different style of a business setup.

[00:21:27] I don't react.

[00:21:29] Eric: When I think IOT, I think of things like connected appliances, hardware, talking to hardware when you use IOT, help me understand a little bit more about why that's a preferable term to use here rather than just saying API. If we're talking about more than just software talking to software.

[00:21:47] Admir: Yeah. IOT means you nail it machine with machine. They should talk and they should give me the result. And based on this result, I do my decision. In theory sounds great. But in the reality, the problem is that not each machine is connected with the automation. So you have different barriers. Either one of your barriers is the language of the machines.

[00:22:09] So not every system is API capable for instance. The other thing is as well. If you have like also security issues, they have clients who say, Oh IOT sounds fantastic but , I don't want to go out with my data. So then we have to neutralize the data somehow to still be able to exchange. But exchange with a lot of data not related.

[00:22:31] So I give an example today. People think that if I take a bill of material upload into the cloud, download information that in the cloud where it uploaded, wherever it is, the information disappears. No, it doesn't. It will always stay in the cloud. And the API technology allows you really to address cell by cell.

[00:22:50] So it's very difficult to track. Yeah. Well, which application is relevant? API is let's say the turbo for the IOT industry or for the IOT technology or for the IOT century where we are right now, because with the API technology, we are able to exchange data in real time on a very easy level. So it's not really, you don't need again, scientists to develop these kinds of things.

[00:23:18] You just have, you just need a thought process and a platform they're different distributors APIs, the data center, like SiliconExpert and plenty of excellent databases they have in their organization. So that is our niche and our role that we say we are exactly in the middle and fulfill the bridges, which the ERP cannot fulfill directly.

[00:23:43] Eric: It strikes me that what we gain with efficiency there frees up the people, working on these things to do the things that only the people can do. The things that the software can't do is that the idea there that we're not replacing humans, we're just liberating them to, to do the jobs that they need to be doing.

[00:24:03] Admir: That's the point, when you have the presentation with the higher management, they see the system, they understand, then you'll go down to the next level. And the first thing, especially when you have a one-on-one meeting, then you see scary faces: "Oh no!" you are presenting now somebody who is replaced, that does not make sense because the sense is not to invest something to replace a human. The sense is to invest in something to grow in your profitability, sales, usually in sales and at the human do what the machine cannot.

[00:24:42] Admir, thanks so much for joining us today.

[00:24:45] Thank you for the really fun and cool conversation.

[00:24:49] I'd like to thank our audience for tuning in and thank Bay-Soft for sponsoring this episode of the SiliconExpert, Intelligent Engine. Tune into new episodes that will delve into more of the electronics industry. Upcoming topics will include how nice to have technologies are now taking a front seat. A deep dive into obsolescence reality and an examination of the ever expanding reach of Edge AI.

[00:25:13] Be sure to share our podcast with your colleagues and friends. You can also sign up to be on our email list to receive updates and the opportunity to provide your input on future topics. Go to SiliconExpert.com/podcast to sign up. Until next time, keep the data flowing.