Firstborn Lessons from Christ

Welcome everyone to this new episode of the European Coatings podcast. We've been growing quite a lot lately and now we've decided to scale up and now we have cameras. I'm not going to be alone today since we are having a little bit of technology here. It's going to be the topic about digitalization and we will have LabV, one of the strongest companies in Europe that we have software providers and formulation support. I'm with my friend Charles Jenink today with us, co-founder of LabV, and we will just be chatting a little bit about the trend that I have been going on, what is about to happen, what he thinks that will be shaping the industry, and yeah, share a little bit with us what he does in his daily basis work. So welcome, Charles, how are you doing? Hi, great. Thank you. Thanks for having me. Well, I mean, it was super great that you managed to have some time for us from your cousin's visit. Anytime. Great. So Charles, before we talk about LabV, do you remember the moment when you realized that we were having a problem with digitalization in the coatings industry? I remember different moments, right? So I'm a mechanical engineer by training, which means I've been trained or studied how to develop products, right? And the first time I was in a lab, it was a very early days of LabV, actually. And I went into a lab and I saw this guy, it was a PhD in chemistry. So, you know, to me, it's like high respect. It's very talented and probably a lot of expertise. And he was just doing copy-pasting in ex-alphys. And I was just wondering why? I don't understand why you're doing that. And you don't seem to enjoy it. So that's where the very first time I think that was that, like the first time where I realized something is not right. I mean, I wouldn't do it. So I'm just assuming the other person wouldn't do it. Unfortunately, I was last week in London and I visited that university there. And I asked, what do they learn in the technical part? And they said to me, they were with MATLAB. They have one lecture that is just Excel. And I asked them, are you planning or thinking to use something with artificial intelligence or software? No, because our students need to learn the basics and the foundations. But I think the foundations have changed, right? So now it's completely different. It is different. The last moment was yesterday. Indeed, I was visiting customer in the chemical industry. Very similar to coatings, but slight difference. But in the lab, they actually use the same methods and the same equipment and everything. And you had this one person. He was, we actually interrupted his work. He was literally doing screenshots from files on the computer, send that over teams to himself. Because two hours later, he goes back to the desk and takes in searches and then takes screenshots that put them in Excel file. And he just said, yeah, that's how I waste time. So when the experts use the word, I waste my time. It's a strong thing to say for an employee in his work context with his boss next to me and everything. They get paid for an amount of hours and they willingly recognize their waste. - They do waste time doing these things. - Correct. So I think it's been a journey seeing over and over different types of problems. But it's actually, I think the very first time was this one person who he started to talk about his job and it was so interesting. And then you see the same person doing something so useless as just copying cells from one file to the other. No, that just doesn't make any sense. But now it went from zero to 100. And now we have software, predictions support, different providers. But let's start with a definition question. So we've heard limbs. What are they? So you can find different definitions. A limb is basically a software for lab information management. It comes from an accredited lab. So everything that is certified and strong from the quality side. At least that's our understanding of it. And we do think this is a good solution in certain type of industries and labs. But we do think most of the R&D labs, especially in the coatings industry, but every type of or every industry related coating. So it could be also adhesives and so on and so forth. I think we think there must be a better approach. So that's why we started to basically look at what exists and what are the current pain points with these solutions. Not only the pain points in the daily work, but you have always the same approach when you're trying to be innovative. You look at how people work or how they do specific things. And then you look at the current solutions and what are the problems with those. So we've been listening basically a lot in the first years. And that's how we came up with our attend intelligence approach. And then you realize that limbs are not covering all the pain points that the industry is covering. Correct. Lab V is a little bit more tailored. So just big to the, so we have in our industry, correct. So do you mean we use some examples? We basically realized limbs, I mean, the first letter is L for lab, right? So it's a lab tool and R&D is more than just a lab. So that's where it started. And we said, how does it look like if the solution that R&D needs is kind of the solution for R&D and for the people in the lab and also for the processing engineers? So that that that was the starting point, I would say, of our approach definition. And then we just asked question like, what are you trying to achieve? And why, what is the tool you may be using now? Because we do have a lot of contacts and now some customers who had limbs or have limbs in transition and transition to our system. And they all said, well, with the limbs, I can almost perfectly manage the process. Everything is documented and archived so that any audit type of work is perfectly traceable and everything. But I don't manage the output, which isn't the innovative material or the innovative recipe. And that's where we thought, but that's the purpose of R&D, right? The R&D is not supposed to be a perfect lab, it is supposed to be a perfect formulation. Before that, you have the production line. R&D should be more like playground where new ideas come up and then you keep everything on track. And that also brings a problem. So our industry, especially coatings, we have a lot of seniors, formulators in R&D is very advanced in age and they are going to retire at some point. Where do you think, and everyone is talking about this knowledge loss? Where do you think lab V is or artificial intelligence support? Is it going to make sure that this loss? Is it going to support it? Is it going to be worse? So unfortunately, we cannot solve this problem that people retire. There is no solution for that. Which is a good thing. It's a good thing they deserve it. So we've been. The observation here is. First, it was in Germany now, it's actually confirmed that it's in the whole Europe. I assume it would be the same in other regions of the world, but we focus here. So there is a whole generation which is somewhere in between, I would say, Europeans and my parents or Europeans. At least my parents retired already. Mine must stay worse. So this generation, the baby boomers, you know, retiring now until the next couple of years. Which is no surprise, but they have been the last ones working mostly before Microsoft, Excel or Microsoft Office was introduced or computers were introduced. So they have a completely different way of gathering information. And we realized that what happens if the person with this expertise being in the department for 30 plus year years leaves? Well, even if it's planned, you don't leave within two weeks. It's like a half a year ahead or something. So you know it. But we've seen customers, we actually had one customer in another industry, but it's very comparable. One of the engineers was retiring. And the first thing HR and the head of R&D did is to offer him a contract as an external consultant part-time after the date he was retiring because they said he's been involved in 50% of our projects over the last 30 years. So it's kind of a process. It's kind of part of the workflow now that we ask him for his opinion and expertise. So that's the loss. That's the problem. Now last year, I heard another one. It's an additional one and it's a different perspective. It's not the perspective of losing. I'm losing you as a colleague. You have so much expertise and I cannot talk to you anymore. I cannot ask you. That's a disadvantage for the team and for me probably as a team member. But if you look at it as the owner of the company, I am basically as an owner investing in people. I'm paying salaries to invest in building knowledge, developing expertise. If the expertise and the knowledge disappears when the person retires or leaves the company, I actually lost my money. And that was a conversation I had with the CEO of a company I come, mid-sized German company last year. I guess it was another surprise, but I never thought about it in this way. A realization point. You say, "Oh, true." That's probably the main reasons. If you want to think a little bit wider, and R&D is the key point to be innovative, and innovation equals probably competitive advantage. So losing knowledge is losing innovation capacity or capability is equals losing competitiveness. I think that's something you've been rather. You had other episodes about big, large movements in the industry. Those parts are not that tough and already. And I mean, we know how difficult it is in Europe right now with the industry. We should just all together search for possibilities. I'm not saying we are the only option and that's the best solution, but we are trying with this software product we have to play a part in saving or keeping knowledge within the company. And building, I used to say, always as a joke, we're trying to build your digital brain within the R&D. We don't call it this way, right? It's also not a digital colleague. It's close to this, but that's. It's a support. Brain support that you can. If you can envision a digital brain or digital memory, that's where we want to go together with the customers. So that's part of the mission, if you. Talking. You mentioned Europe and that we have already a lot of. MarketMovs. How do you see Europe in terms of regulation, affecting rich. So we have also the topic of PFLAZ. Low VOC, we have every month comes a new problem with our chemical component. How do you think this is affecting the need for a structure data or the need of these kind of software? I think it's just increasing the need. Because you need to know more, you need to be able to react very quick. Take an example. Tomorrow you have a new raw material that is forbidden and you have to replace it in 10, 20, 50 different formulations. You can do it by hand or you could do it in a more structured way. You could even think further. It's already possible. If you have enough data, you could almost use the software to actually make replacement suggestions. It probably has negative sides to have so much regulation, also many different regulations. I'm not deep enough on the chemical side to evaluate that. I just hear or listen to customers and I think that's difficult. It's probably a good thing as a consumer. There is many sides of it. But if I look at the work in the R&D, it just puts even more pressure. It's highlighting the need for speed, reaction time. I have to be reactive, which means I have to know very quick what to do. Where is raw material in which formulation? Do I have an alternative? Do I have to qualify in the raw material very quickly? Which is also a topic. There is an example. When you go to a fast food, the french fries in this packaging, there is a coating around it. There is an American fast food chain. They have, for example, in Europe, one very simple requirement. The supplier of this packaging has, in case of a request, about a charge number, like a batch number, whatever it is, they have four to six hours. I think if I remember correctly, to provide all the information about everything, including the coatings, like the everything that was on the packaging. So imagine if you, you know, any type of regulation is just interesting. That you say we have a lot, there is a lot of demand for speed, and then it's a new regulation. With speed comes competition. And here in Europe, we have, of course, you, Labille, and we're very strong here. But we have, we've seen it also when companies, that quite other companies, here in Europe, America, China, India, and so on. How do you see the competition in this new field? Because this is something that just has been going on for five, 10 years maximum. How do you see the other companies' approach in this more? Like, we try to keep it in Europe. The other companies trying to come to us as well already here. So the main competitor is Microsoft Excel, to be honest, right? If you look at the software providers, of course, LIMS providers, there are a lot of them in Europe. Many of them are very good and well-established. It is a type of competition in Europe. There I do see some movement. The realization that it needs a little bit more focus on the data, maybe a little more work on the AI possibilities, whoever, you know, there are enough possibilities to integrate that. And then you have basically a couple of software solutions very specialized, and then there is us. And everything else is actually American based. Or somewhere in Asia, Singapore and so on. But it's a lot coming from the US. And the interesting part is, they are huge companies. They have large funding. It's a different type of competition. Right. It's a different league. On the other hand, they don't care so much, I think, for SMBs. And so our typical European SMBs to German middle stance, these, these layer, the layer 100 to a couple of 100 million euros, they struggle a little bit more as American companies, because the decision process are different than they used to. If you go to a large one, I don't know, the Henkel, Hempel, a BSF and so on, when they purchase a solution that's a well-established process, you know, there is this screening and companies and everything. And you do demos. It's not very difficult to sell. You don't have, you just have to be the best. The layer under that, it's a little bit different. It's a little bit more into understanding the customers. So we just looked at that. We looked at ourselves. We are a European founder's team. I'm French and my co-founder at Steffan is German. We are a German company. We belong to the Nets Group, which is a German family on business. And we are truly European. So we said, well, let's make Europe our playground, because our mission is just to help European companies and the European industry to redefine how they do marketing and innovation, because we think the European industry, especially coatings industry, which is live everywhere in our daily life, has been the best industry in the world. We invented so much and so many things. We don't want that to disappear. So we want to just be part of this and play a role in that. I think it bit. Yeah. And talking about customers, one of the problems we get, so when I hang out in another conference and chat with from paying manufacturers, they say to me that now that they have so many options, that means a lot of different generated data. And since it's not created by a lab scientist, it's created by a machine. They don't know to which knowledge does it belong. It belongs to the company, belongs to the software provider company. Where do we stand there? We have a very simple approach. The data belongs to the customer. We own the software. Our job is to software. So our job is providing the tool to be innovative, to develop the best simulations and so on and so forth. Whatever is in the database, it doesn't belong to me. It belongs to the customer. And whatever is generated by our AI assistant, or I don't want to say generated, but maybe any new learning or potential IP coming out of that, belongs to the customer. That's how we think. We don't have any trouble getting information from lab equipment into our database, because at least not that I know, but all the manufacturers they don't use, they use basic data formats like CSV, Excel files, or you can use it. It's not encoded. I've never heard about someone complaining about the manufacturer saying, "No, no, what you measure is mine." So I do not see a large problem there, but it's a question that is asked, like it's raised, there raised the question almost every single time. Is it deleted now? Is it lab-based data? Is it my data? Yeah, does this. Do you have access to. So do as a provider, do you have access to the data, your customers give into the system, or is a completely closed system that they can use or bled at the needs of their whole database? So we don't have access with one exception, which is our technical support in case of a technical problem, they could look into the customer's data system or database. I've seen my customers data only when visiting them. So paying a visit is helpful because then I actually see how it looks like, right? Sure. But other than that, we have no access. We don't see anything. We do have very simple things we look at, but it's completely. It's all over the customer base. It's just our numbers that we look at, but that's it. I don't know what they're doing on the formulation. I don't know which raw materials they use. I have no idea. And we don't want that. First of all, we didn't see any value, because it would not work to build a large data pool from the coating industry, because everyone is doing something different. You cannot compare a paint formulation for automotive application with something with outdoor product or IKEA furniture. Like what's the point? So we said, we thought about it very early. Is there any potential for the customers to benefit from that? We didn't find anything, so we said, we're not going to work on that. We work. It's very important to us, and I think every provider is doing the same, but we worked very focused on how to make the customer's data as safe as possible. Every single database is separated from the others. Like there is no chance another customer can go into your system. There is no chance our team goes into your system. There is probably no chance that anyone from the outside goes into your system. There is no chance, because it's import technically impossible, that some type of a stake exulfile through hacking is being sent to the database in your name. It's not possible. That we have the technical, let's say, technical infrastructure in place to do it. There is always this concern that the system is going to learn from the input you give, and then the company behind could harvest that to tailor or approach different companies with their product. So it's this concern of, it's my data just being used by myself or by my team. And I think that it's a good point, because it's still new. Thank you, Chad G.P.T. It's like you use Chad G.P.T. and you actually have to tell the system if you want your data to be used or not. In order case it's completely different. The AI assistant is not learning from the data. It's an LLM mapped using the database, but it doesn't learn from the data. It just uses the data, which is different. We don't train the LLM, which is used as a standard. Best performing LLM right now. And it has access to one database, and it can do a lot with that. If you think about predictions and so on, it's different, because it's a proactive training when you use machine learning. And that we only do if the customer asks for it. And we usually do it with the customer anyway, because we need a human expertise. So I don't see a risk, at least not the way we approach it. I cannot speak for other companies. Everyone is different at the end of the day. But back to you to LabV, now it's going to be used in Estlingen. So you're going to have collaboration or a gig with them in the university. How significant is this a step for the industry? That now the future formulators are going to learn how to use a software. Again, a type of realization, if you want, but what is the transformation for an industry? It's what happens in a company, and it's what happens in the educational side. And for the first years, we were in touch, but we'd never tackle that very seriously, I would say. And when we came across Professor Shachmann, who is probably the one who is teaching-- I don't know, I mean, he presents 70, 80% plus. If the German or German speaking lab experts, we both saw that there is an opportunity to just show students and tell them that there is a different way to work. Doesn't have to be excellent chaos. It can be something different. So starting this new semester, March in March, 2026, now the students who are working in the lab are going to use LabV to document their experiments, document their formulations, the variation between the formulation and what they did in the lab, and the formation in the sample. And probably use LabV to, as an input, to their project reports. So they compare different type of pain formulations, different temperatures, and so on. What do we don't do? Yet we kind of deactivate some of the features we have. So the AI assistants will probably not be used very much at the beginning. And the whole reporting, automated reporting, functionalities either. They should learn to do it by themselves as a correct. That's what we don't want to-- we are engineers as well. We have to try and learn to do mistakes. So we don't want to interfere in that. And whenever the professor feels that it's the right time, or feels ready for it, we will follow. We're not pushing very much. What we will do though is maybe this semester, not sure yet. So it could be next semester, at least having kind of a keynote, or a course, maybe an hour, to explain to the students. Restatio. Yeah, so here, what is about AI in R&D and in coatings R&D? So that's kind of the approach, step by step. But that clarifies that it's a change. And now there is not Excel chaos. Now there is a new tool. I remember when I studied chemistry in the university that I had like three months of Excel, three months of MATLAB, and three months of Wolfgang. And now it seems that all summed up, and now LabV will be the thing that students will be learning. But this comes with a risk, or at least professionals can see a risk. Are the students going to relay heavily on AI or supported formulation and take away fundamental understanding or creative concept that at the end helps to do new simulations or improvements? I think this is going to change, like the way they learn, is going to change in the future. But it's not just in this specific case, it's going to be everywhere. I would love to make an interview at the, you know, after, I don't know, somewhere in the university, interview the students and ask them, like, how much of your studying you do with chat GPT? 80% minimum. So I think that's going to change. It's for sure. How is it going to change? I'm not sure, or I don't know. I don't know. I don't really-- I do have ideas. We work on things. But basically, for now, it's about informing them, telling them what's possible, and maybe being this kind of translator between technology and the students, because I don't think they have any module or whatever about AI or not so much. And maybe step-by-step is going to be the same way you described for your-- in your case, you're going to have a four-one semester, a two hours a week AI course, and then you're going to learn everything about prediction. But I can imagine that the students in the future are going-- will basically go into the industry, leave with their masters or whatever it is, and then have a different approach to their new job. I'm convinced about that, because there is absolutely no reason to keep doing this Excel-based R&D work. It's not working. And funny enough, it's very younger people. They actually don't understand it. I can barely use Excel. It's just like-- it's very early in the early, early day of the day. We had a young person. There was no student anymore, but a young person, like just young professional. And the person actually told, you know, when you visit a customer, there is the official part and there is a part where you have lunch or something. They're a little bit more relaxed, right? And obviously, you know, we were talking a little bit of our life and everything. And he said, I arrived here. They had paper. I'm sorry, but paper. I don't even know. I can write, but I never write. And I think that's the same thing with Excel in AI. It's not at some point. It's going to be, what do you mean with Excel size? What's that? I don't understand it. Yeah. It's like telling a 15-year-old, you know, you don't have to send a 15-minute voice message just call the person. They say, what do you mean with calling? I don't do phone. What is phone? I think that's-- So you can apply this to pretty much every hour. Yeah, every side of what we do. And to keep now the vision in the future, Labvi is very strong in Europe. I would say potentially top one we will have in. So where do you see Europe? Is it leading or lagging in digital air and the infrastructure? For what do you see from your customers? Depends on what should compare to what or which region. Compared to the big players. Compared to the US, we are probably behind. That's probably that. And there is also a different mindset, different approach to new technologies. But we're not-- I don't think we are as far behind as the whole Europe. like Europe, Beijing, Figu on LinkedIn and these kind of things, it's kind of depressing sometimes. I don't think that's that bad. Companies try a lot with the resources they have. That's the fact that they want to motivate it. So I think Europe is catching up with its own base and also European style. And I think we have to accept that. We are not Americans. If we were Americans, we would be in the US. We are Europeans. We think different. We're not the country. We have a union of 27 countries. We are 27 languages almost and 27 cultures and 27 different goals as well. In the US, it's a little bit different. It's also a federal union, but it's so I wouldn't try to compare so much. I would just try to get inspired by other regions, even China. China is probably so much more innovative than the US in some aspects. We just don't look at it and see how can I use this best practice and translate that, adapt that to my European situation. So it is behind, but I don't think we're so far away, especially not in the cutting industry. It helps to listen that we are not. It's always the typical thing that in Europe we bring the regulation and then the rest of the world makes the innovation. But I think in this case, we are also taking part into the game. But we are going to sum up a little bit the podcast now. And I would like to ask you, what are the next strategic steps for Labi? So what's coming next? More and the better use of artificial intelligence functionalities. We're not in AI solution. We are software solution with AI inside. So prediction as a one word is the word is prediction. And then the prediction, it goes from predicting or planning potential experiments up to at some point formulation prediction, which is much more complicated. So I think that's that's it. Everything as it's just my nor productivity, improvements with I want to say daily, daily functionality is like smart things that people actually want to solve. But the big big big big developments for this this year and next year is definitely coming from what we call the AI assistant to a co engineer or co chemist. That's that's the vision within agent system in in the software that is basically fully integrated and not, you know, so that's where we put the most the largest effort right now. And now like if you were running for president thing is for our European and the John managers of future experts, one piece of advice that you would give them about software artificial intelligence new tools. I think in coming back to this question about Europe versus others, for example, the US, I think that's this type of mindset. The risk and versity is is not a great thing. Obviously I'm biased as an entrepreneur because I just live a risky life, right? That's that's the core of it. But if you know, if it was only one advice, it's just it's not software doesn't mean it project doesn't mean complexity doesn't necessarily mean big risks. Just have a look at it as for demos like challenge the software providers. Don't let software providers convince you that, you know, if they say yes to everything, there is something wrong. So be the one who's in in the position of challenging and say, I want this. Can you do it? Yes. No. Can I try it? Is there any way for me to try it? And if the provider is is good enough, it's going to help you. And I think it's true for everything ready to digitalization. I was taking a little bit the example of levy, but just be ready to be a little bit more bold. The small risks if you want to call it risk, but you're not going to lose anything, you're going to learn a lot. And that's the only way to get better. Before it was used to be in the laboratory, you would challenge an experiment, try it different and now you just do it with software. So that's what I'm still struggling with. The people in the lab, even the people that the R&D experts, the engineers, the chemists and everything, they do trial and error every single day. But somehow when it comes to larger decisions, they don't know how to do it. So it's weird. So yeah, if I would move in person of advice, maybe that's a better summer. You're right. You select what you do in a lab to apply to any other decision as a manager. Yeah, try an error. That's the only way to learn. Well, but with this, thank you so much for today. It was your five family. We could be talking for two hours more. Probably. And thank you so much. All of you for listening and we see each other for the next episode.