Process Mining as a Commodity with Gia Thi Nguyen, SAP Signavio

>> The mining your business podcast is back yet again with another episode. How are you doing today, Jakob? >> I'm doing very nice, Patrick. Thank you. >> Our guest is no stranger to sharing his opinion and content online. We are lucky to have him on a show today. Yati Nguyen, VP at SAP Signavio, joins us on the podcast to talk all about the suite of process mining, process mining as a commodity, the importance of a hero in the transformation process, and what gardening has to do with business process management. You ready, Jakob? >> Hell yeah. Let's get to it. [MUSIC] >> Today, we won't be serving you, Cap of Tea, but a full-blown fruitful discussion with Yati Nguyen about your favorite topic process mining. Tea has a long and decorated story with Siemens, where he worked as a head of operational excellence, which surprise, surprise, is focused on process mining. And who now serves as a vice president in SAP Signavio? I think Tea brings on the table way more than we can actually drink in one session. Tea, welcome to our show. >> Thanks a lot for having me, guys. >> I hope you don't mind my reference to your LinkedIn hashtag, Tea with Tea, and I thought it would be actually a funny intro. So, I hope you don't know. >> Tea anyway, you essentially went from a raving customer of process mining tools, often promoting and praising the technology, while building up the fastest growing process mining implementation in Siemens back in the days. The leading role where you kind of help shape the new generation of the tool in SAP Signavio. You've been basically writing your new chapter as a VP in SAP Signavio since December 2021, and my question is, what is it about? >> Yes, so let me try to break it down in a couple of things. So it is absolutely true that I'm really, really fascinated by process mining. And I'm really privileged that I was also able to start very, very early with this topic in 2015. But if you actually have followed a little bit, you also know that the process mining is just like a tool and a way to achieve certain outcomes. And also, my professional background in Siemens were also very, very, let's say diverse, you know, like it started in IT. And then I became a CEO, then data, head of operation, accents, and even data. I went into service business. So really, for me, the most fascinating thing obviously is actually working together with people of diverse backgrounds. So it is really more about the outcomes which we can achieve with different technologies. Obviously, not only process mining, but lately, of course, process mining is so many people say, like when you get on the blog, I guess. >> So Tee, what is it that you actually are doing now as a vice president in SAP Signavios? This is something we usually like to ask because, you know, for us, we know what data science means. We are also, we have IT backgrounds. But when we see these positions, it's very difficult to actually picture anything for us. And I bet that there are also a lot of listeners who, when they see a vice president of Central Sexcellent, what does it actually mean? And what are your daily tasks today? >> Yes. And it's absolutely, absolutely a valid question. Even a CFO of one industry might have totally different topics to do with another one. So, as I mentioned, I used to be a CFO. I talked to my CFO friends. If you are a CFO in venture capital, it's totally different if you are a CFO in discrete manufacturing. So the center of excellence, what we do primarily in SAP Signavios, we are actually globally set up. So we really have one also in the Americas. We have one in Europe and one Asia-Pacific. So I am part of the Asia-Pacific one. Generally speaking, we support the go-to-market activities. So really, whether it is a creating content and delivery and doing enablement for all the different go-to-market functions, yeah, really from resale sales. And services, consulting all of these go-to-market activities. And then we also do the evangelism and the fort leadership related to what we want to achieve in SAP Signavios. >> So, talk me through a little bit more about the details of this. If you are in pre-sales and then also go to market and all these things, what does your day-to-day look like? Is it a lot of talking to clients, customers, existing customers, new clients potentially? What does your day-to-day look like? >> Yes. And obviously, I mean, but it is not only to my work, but every one stays pretty much nowadays, very diverse, I guess, even for let's say field sales, obviously. So basically, my primary function is not to have, let's say, per se, always the direct customer contact. So basically, I enable our internal go-to-market organization. But it is true that I do, let's say, calls and calls to the someone dance with specific strategic customers when they do need address the very relevant topics of, if I want to implement a tool, I know who to talk to. But how do I make sure that I realize the value and do the change management? How could I actually also set up an internal center of excellence? Do I need to hire 20 more people? How can I go about these kind of change management topics on the generic way? And obviously, there's luckily or unfortunately not the Google recipe to it yet. And we also work obviously with our partners, which are also, let's say, the consulting companies, right? And then we basically provide, obviously, not only the technology, but we provide, you need the methodology, the content, the ecosystem and the technology together. >> Yeah, speaking about the content, I noticed that you are producing a very interesting, like I would say, almost a show or sequence of videos on LinkedIn, which you, you know, it was already mentioned. This hashtag T with T, what is this about? I like you are talking about various topics there. Almost seems like you're going in this direction of helping out your customers, or generally everyone who's adopting this technology in a broad scope of LinkedIn. >> Yes, so basically, really, I've been quite different social media or on LinkedIn. I think also really when I started to drive the other management for tomorrow program, which was probably around to 15 to 16. And basically, what I really, if you follow me, I very seldom share links. Because I think that it's 99% is actually always own content. And I'm not saying that all of this wisdom came purely from me, but I try to rephrase certain oddities, strangest, what we see. And that's why I try to talk about transformation, why is it so hard? I also talked about resistance to change, and this is to the last episode. Honestly, I have to actually disagree on many topics, but it doesn't mean that it would be wrong. Because again, I did not buy or own the wisdom here. And I think I try to offer, let's say, a different kind of point of view. And the fact can just pitch for a couple of seconds, like my first episode was actually called complexity. And I hear, especially, other IT software vendors and ourselves, including, we talk about complexity being, I'd say, the silent killer of performance and things like that. And I totally disagree, because complexity is something which is natural. A human body is very complex, a jungle is complex. I think we need to differentiate to waste. I don't like waste, but complexity, that's nature. So that's why I try to offer a kind of differentiated nuance point of view on certain things. Maybe as in your next video on LinkedIn, you should just pitch our episode and say, "Then things that are wrong about these guys are saying." I said I disagree, I didn't say it's wrong. I think it's actually valid and I really enjoy and also support that we do need to have, let's say, more own content, own voices in this space. Because there's one slide that I can just picture to you. I always bring up and it says, basically, why do transformations fail? And one is, let's say, from a Kinsey, one is from Harvard Business Review. And I ask the audience, because most of them actually have seen it. And I ask them, "Do you know what, if you read it, what is the only difference?" And the only difference is 24 years. So how can it be that changes the only constant in life and still the best paid researchers write the same BS for so long? I think this is really important to understand, because apparently, we are not having the right discussion. And it's the same like on LinkedIn. We have been clapping for the nurses for two years on our balcony to save them. I have been in healthcare. In two years, I've seen more dead people than in all my life before. Yeah, I wasn't Vietnam. And do you think any nurses earning more money now? None of them. So I really do believe that all of this spiel about transformation is really a very shallow talk in many cases. What I think is always fascinating is when people get into creating their own content and sharing it online is kind of figuring out their motivation, because I know why we wanted to start the podcast. We wanted to say, "Okay, this hasn't been done. Let's just go and talk about it, because we think it's interesting." What was your motivation going into starting this TV team? Actually, I was already on social media, as I mentioned, but it's true. I had a hiatus I was out, and the reason was actually, and it's a little bit ironic, exactly the reason why I was out brought me back in, which was the noise. And I could not take that noise anymore, where I said, "You know what? Even if no one listens to me, it's fine, but I cannot let that stupidness noise be there, because I have seen it differently." And my own personal motivation, generally speaking, why I share all of these things, is very easy, because I've been to hell, and I'm lucky and privileged to have survived it, and I don't want anyone else to come after me to make the same painful learning experiences. I wish that they have satisfying learning experiences, and then overpass me in less amount of time. It took me 20 years to accumulate this kind of wisdom, if they can do it in two months, and are going to be my boss, man, I would be so proud and happy. Sounds amazing. I mean, the more learning we can get, the faster we all grow. So I also want to just mention that I love the diversity of opinions, and I mean, one of the reasons why I'm so excited to talk to you today is exactly that, that I know that sometimes we don't necessarily agree on things, and I think it's just beautiful, because it brings just a whole new dimension to the conversation. However, now getting back a little bit to those process mining topics, as we still have to satisfy our customers. Yeah, well, our dear listeners. So just going a couple of years back, what was your first exposure to this process mining problematics? The thing is, for me, was really absolutely eye-opening. I couldn't even imagine that I could answer questions I didn't even dare to ask it. So, in Siemens, we were looking at over 400 million different events, and each event had like 200 different columns. So literally speaking, I would know each and every price change on a level of a material on what day by what organization. And that was really mind-boggling, and I could even know what was the value before and after. And then, of course, it really fuels the whole use cases around the deep learning and machine learning, and all of these other beautiful things. But very quickly, and again, it's all very well documented, so I'm actually just repeating memory data a bit. But very quickly, I understood that this is a farce, because everyone wants to have transparency, but actually it's not really true. What they want is a fast transparency for which we don't have to pay for, and also know what to do next. So when I say I want transparency, it's actually three things I want. And the thing is that I had a, what I always used to say, there was a transparency beyond imagination, and I had it exactly wrong. It was imagination goes beyond any transparency. And I think really it sounds ironic, but hear me out, we are losing our humanity, and we become more like robots. So that's why I'm not afraid of robots taking away my work. It will not happen. But for me to rediscover my humanity, which means compassion, which means empathy, I do need technology. And that's why I was also doing a lot of key notes where I really said that technology is exactly like art, like music, like a sculpture, a form of human expression. And that's why we should not actually utilize the terms like human versus technology. We should talk about humans versus humans who created bad technology. And that is a really, for me, it's a very, very different way. And therefore don't talk about unreliable data or trustworthy AI. What did they do to you to be put into the corner? Who created the, who created the data, who fed the data, who did the algorithm, right? And I think this is really the discussion we should have. So I'm not talking so much anymore about the why and the what. I'm actually more assuming that some why will be found for whatever case it is. And the question is really how, how in terms of quick and inclusive, how can we do that? That is basically what. And if the technology process mining has a certain capability to do so, then all for it. If it's going to be IRPA, X, I, X, whatever RP, they call it nowadays, I don't care. Then we can do it. What is in your opinion, the specific edge that process mining as a whole, as a discipline, brings on the table in solving these things and in bringing what you were actually just saying. Yeah, it's actually not really creating something new, but it's basically rediscovering what the organizations always have. So it is a democratization of the knowledge, which was there, but it was very hard to communicate. So using process mining, whether it is a visualization or whether it's a data driven, it basically shows what people, quotes and quotes, in many cases, knew already and have been talking about, but nobody wanted to listen because they didn't have the new sexy dashboards. But now, they become basically the creator of a content themselves. There's a certain democratization. And then it also spurts the creativity to ask better questions. I'm a person I never care about the answers, because I believe if you ask better questions, you would come to the right answers together anyway. And that is really where process mining was helping quite a lot. On a technical functional level, it's also a very simplistic technology, if you think about like an entity relationship diagram with a very. Yeah. A primary key. That's simple by it like that. And it's beautiful. Why is it beautiful? Because simple is a very hard thing to do. And making it then accessible to all levels of organization is something which is not every, let's say, not every technology has. So everyone can say the word AI. And if I ask, what's the difference between AI, ML, and DL? Then already the discussion is a little bit different. And then I ask, what is the difference of a gradient boost in terms of algorithm and AI context? Then the population goes even closer smaller. And if I talk about convolution and neural networks, it gets even smaller, right? And that's why process mining is much more accessible, I would say. There are still some hurdles, I think, in how people interact with process mining because you know they see, okay, we have four million cases, here's a graph. Okay, cool. Now what do I do? There's that initial just hurdle to get over, okay, what am I looking at? What am I on at its core, at the reality, looking at what this tool is telling me, right? Do you have any tips or in your experience, what has been like the go-to thing that you tell people to look out for for them to get a better grasp of this new process mining thing that they're looking at? Then let me try two, maybe three angles. One topic is a little bit, let's say, functional technical where people always ask, like, what's the difference to be I? I think it's a valid topic when you talk about, let's say, architecture and you talk about, let's say, certain use cases where something is better than the other, yeah, arguably. But I don't think that this kind of discussion or the answer, therefore, will lead any possible valuable outcome based on that. It's more a philosophical question, so to speak. The topic of how to grasp the case event blocks, attributes, and whatever language they're called, I think what is much more important is to talk in pictures. People do understand. So, for example, if you think about the process mining in the view, I always talk about that's our train stations. And if I want to go from Norway to Italy, I don't need to see every small village. But if I want to see every village, I can zoom in and that's all the different train stations I have. By the way, there are also different kinds of trains. So the arrows are different trains. There is the fast train, the night train. There's a train which only comes two times a day, you know? And of course, some trains you go back and you can go forward and you need to change the train and things like that. So that's one metaphor I consistently use. The second topic which I believe is also very hard for some people to grasp is, and I think the Americans would say, "You can always pull a rope, but you cannot push a rope." So this is actually very easy. If I explain to someone, I have a thousand quotations and not every quotation will become a sales order, and not every sales order has been invoiced yet. And suddenly, if I want to see the invoices, where are all my quotations? I said, "No, no, no. It doesn't work that way. Who is your master of a universe?" And that are very logical questions. It has nothing to do with technology or process mining. And so it is something where I believe when people talk about, let's call it a simplified 2D table, and then you join them. Some people have not that, let's say, literacy or experience on how to ask that kind of data question. And with this knowledge of how to ask the right questions, obviously comes this transparency that you spoke of earlier. The transparency also can be, as you said, a little bit of a negative because you see everything, but it doesn't provide you solutions or the answers to the questions. You can just ask. It kind of says something. It just says, "Displace data," but it doesn't really tell you what to do or how to then take it and then produce something that will reduce cost to whatever it is. Is there some tool in your opinion or do the tools generally lack the next step? How to go from, "Okay, here's your data. Here's the problem. Now do this. Is this a general tool issue or is this more of an understanding issue or what is the gap that is still missing so we can easily go from? Here's the problem. Here's the solution. Okay. And you cut me off if I get too pitchy, but I will not say a specific name, so I think it's okay. No, okay. So even if you just do a Google search on my name, you see that I've been friendly with a lot of different IT vendors on a larger scale, right? And the truth is I was very successful implementing workflow automation, robotics, process automation, process mining. But this is a story you can find on Google and the nice YouTube videos in the keynotes. But you don't see so much, but I'm very open to talk about it. Is what about the missed opportunities? The missed opportunities were closed as good as possible by the people working. And the thing is, the reason why I met SAP Signavio is absolutely no spin on marketing or because they provided me with some golden package. No, because it is exactly what I have been struggling with in the career. If I ask you, you please make the most beautiful garden. You don't even know whether I want to have fruit or have only flowers or vegetables as well or all of it. So you don't go there with a just a shovel. You need to have different tools. But it doesn't mean that I tell you which tools you must use at what time you want to have a shed and what you need at what circumstance you want to utilize. And don't create your own shovel and invent it by yourself. So therefore, in SAP Signavio, we have this holistic view and therefore we do modeling. We don't shame modeling just because it's old. We do process mining. But do you know what process mining and modeling if you put them together bring different topics of non-conformance and understanding what can be a digital twin? And that is basically what we believe here in SAP Signavio is to do the holistic, but it doesn't mean that we shove it into the throat, that you have to use it all to realize that you can use any different journey because we understand that every customer's journey is indeed unique. Are there patterns? Yes, but the journey is still unique. Well, to be honest, you are not the first one who is framing process mining in this, I would say, relation as we also had an episode which we literally called that process mining is a hipster child of business process management, which was also eyes opening because hearing these experts who've been in the field for quite a while and now seeing this emerging technology that everybody just jumps on completely forgetting what it's actually supposed to support, which ultimately is a better process. And including all these tools together and just going hand in hand is also more and more in my opinion a way to go to have this, as you mentioned, the holistic view of the whole process. So is this one of your footprints that you're now also incorporating into where this SAP Signavio is going? Yes, exactly. I mean, if you ask me, are the features and functions there necessary to really embody our vision yet? No, obviously not, right? But we really have a clear strategy and vision where we want to be. We also do believe that faster insights and also faster time to adopt is really the name of a game because this is what our customers want. It's not what we want. It's really what our customers want. And I think also the certain empathy and compassion inside that quotes and quotes, that's a cold technology. It's actually very important. In other words, the suite which we provide is a very integral part. It's not just some side note or some afterthought. No, when we talk about our collaboration hub, the social aspect of driving the transformation, it's that serious. So that's also one part which we really push to say, sharing is caring, but it also gives you a real big benefits. That's really important. Actually, people were in front, or I would say upfront in many of your previous, I would say, pitches with different vendors where you really brought attention to the people who were really driving these initiatives, be it, you know, especially actually in Siemens, where I remember you had a team who really, really made the whole initiative grow. And my question would be, how important these people who are eventually driving these new technologies be it to say, be significant, or any other process mounting, or RPA vendor are, how vital they are in these initiatives to actually make this success for the companies. And, you know, having this mindset of change and being able to also share it with other people. I mean, absolutely vital, I mean, that didn't require much expertise from my side, I guess, to say that. No, but let's go a little bit deeper onto that question, which is a very, very valid and relevant question. I think what we need to understand is, and I always quote the paraphrase from the Pixar Disney movie Ratatouille, where the red becomes a copyright. What I always want to say is, like, not everyone can be a digital hero transformation agent, but the transformation agent can come from anywhere. When I was a CFO and I had a really hard time how to deal with this very financial topic, I got help from HR. What I'm trying to say is, if we actually think, who are you to give me input, you are HR, you're not finance, then we are limiting ourselves. And that is actually what's happening with the digital transformation. We need to be open minded that anywhere that hero can come from. And therefore, we need to make sure that not their bystanders and agents, but everyone, to some extent, becomes a contributor by themselves. I think that is really important. You mentioned then that in SAP Signavier, this, you know, people play a vital role in how you approach the customers and that you are trying to enable them and everything. Is there maybe some specific strategy or specific way on how you create these heroes, create these technological heroes to try and. A magic wand perhaps. Actually, honestly, it's super easy, but I know that the execution is extremely hard. And this is actually also something I have talked quite a lot about and I actually have done with the technology research companies, the white paper on. And basically, in some cultures, they call it the middle-out approach. I call it the middle ground. And what I want to say is, is it the top-down or is it the bottom-up? And the answer is clearly it's both, but not the same. So what is the nuance here? So you have to do the top-down in terms of the context. It's the trust, the resources, the time. If top-down, you say, let's do nine and not alterings, you lose. If they enter into that discussion, you lose. And you do bottom-up the content. And then you will deliver faster with less resources and less time. However, where does the bottom-up and the top-down meet? And that is the middle ground, because in my project, I had people who actually can walk the top and also talk the top seamlessly, being authentic on the shop floor, but having no, let's say, fear to speak the truth to the executive management. And yes, it's true. I still know what is RV60 SBAT ABAP. Because that's what I used to do in 20 years ago. And that is important, because you cannot talk about the automation of billing. And we need to do that in our O2C. And then in your entire organization, no one knows what RV60 SBAT is. It's impossible. So what I'm trying to say here is that identification of those heroes is simply a superpower. Extremely easy to do, but very hard to be doing consistently is listening. You need to just listen to your organization. By the way, sorry if I'd, and then I would be quiet, that's why in your resistance to change, I really would have hoped that the resistance to change is actually your gold mine. That's a diamonds. Yes, there are some party poopers there. I agree. But the resistance to change is not the same natural resistance to change, which is biological necessary. That also exists. I think that the resistance to change in organizations, which you see is 29% people who do care, but have given up because the culture has become inconsistent with what actually is expected of them. So they are only resistant to change as a protection of the corporate culture. So therefore, they are actually the ones who are your SBAT for transformation heroes. Go to the person who complains the most and make him the hero because do you say, now I listen, now you get the money and I'm sorry, it took so long. Very nice. I have seen the sentiment with a lot of clients where some people, whether it be from the sales or purchasing or whatever, just loves the tool and loves the analysis that we've been that we're building and just takes up the helmets and says, okay, I'm going to focus like 20% of my time now on what I used to do and just do 80% now just the data science and blah, blah, blah, and the best of managers will just say, okay, here you go, go do, right? You've been with us for 20 years. You can know everything, just go do and go build some stuff, right? And it's that freedom that really makes those diamonds, as you said, shine, I think. But what I was going to ask because you mentioned it briefly, is this social hub thing in Signavio, I'm starting to figure out how that looks and what it enables users to do, I think that's it. Yeah, so basically it's actually quite a simple, and again, it's a hard thing to do. It's really think of it as like your place where you can consume and share the process knowledge. It's right here, a single source of process truth. So there's the dictionaries, you know, when you'd say, for example, approval, there's a hundred thousand things you can approve, right? What does it mean to approve this document? Does it mean I approve the financial aspects of it or do I approve the vendor due diligence has been done or whatever. And all of these things are hundreds and thousand times recreated and suddenly we talk the same language, but we mean different things, right? So the collaboration hub is really the space where the people can meet, share the content in order to not reinvent the wheel, but also to accelerate to not start from stretch to get started and have this exchange. And I was close and close kind of rebuilding that collaboration have back in 2016 by actually utilizing something like a, let's call it a Facebook/Twitter/SharePoint, somehow kind of put together and just we have a concrete example so you know what I'm talking about is we had the policy of zero manuals. We just said it's a Vicky and an FAQ will turn into a manual by the people. So suddenly I had two thousand people, I don't know who they were and they were actually creating the manual of my tool. And I just had to do some certain moderation to go in there to check, but it just went there. So the adopting the acceptance of the tool and also the build-up of the enablement was really really so fast I cannot even imagine, yeah, but it happened, right? So so I think this is what true collaboration is about. And collaboration is not really this thing where we all have to feel good and hopefully at the worst case we do a compromise. No, collaboration actually means to be caring about people even that you disagree, right? And I think that is really the hard part to do, but if you do it together and it is transparent then there's a chance. So it's kind of scratching at the surface of knowledge management in a process focus if I'm understanding it correctly, right? See the thing is that and again I don't want to go too much into semantics because I actually did write a bachelor out knowledge management when I was studying the last millennium. And I think in Latin you would say no man is the old man, right? So I think I do understand where you're coming from and also the knowledge management is really important. What I think is that it is also about because you know that there is not such thing as like to say just process, right? When we talk about business process management, you know, I understand we talk about models and things like that, but it is basically a lot of the mindset of do I share and do I actually accelerate and de-risk the chances of my errors, right? And but processes whether they're good or bad, they also need to change, right? So I think this is also not only about the good model, the good execution because that's always static. But we need to also look at a meta level on how do I change actually also my change management there? Now speaking very concretely, since you are now co-creating where SAP Synavia is going, where do you actually want to push it looking forward to 1, 3, 5 years maybe? I think and actually it's actually a little sad because now it's over 5 years already, so I don't know whether I was too impatient or I was a bit too optimistic on that. I do believe that actually in terms of technology we don't need the next new big thing, but what we now need is the convergence of different things put together, right? So when we took a look off, let's say process mining, entering a graph database approach, right? When I need to actually use a graph database to fuel my deep learning machine learning in order to have better models generated to create bots, right? So I think it's really the convergence of different things put together. And it's also that in the beginning there was a lot of fear when you do about automation and you lose jobs, but actually if you look into the shop floor, the the co-bots, right? The collaborative robots, it's actually the way to go, right? And I think this is also, right? You know, we talk also Siri and Alexa, why don't we just say, "Hey bot, please automate the purchase orders?" And I do not believe that saying, "No, if you have a really good execution management system, you don't even need to tell the bot to do that because we do it immediately." But things change, you know? And I think we have seen during the pandemic that an extremely efficient supply chain is extremely brittle, and then a lot of, let's say, quote-unquote, "less matured economies with more manual processes were much more resilient because nothing went against the wall and they did it manually, so they were actually more resilient." Right. Just hearing this, this, "Hey bot, automate something is already bringing so many wrinkles on my forehead because I know how much work has to go behind these things." However, what I also wanted to ask, there is a big discussion also in process mining communities about this modularity of process mining that you could basically, you know, make the tools itself kind of replaceable and just play with the technology itself. And it can even go a step further where I heard that certain companies, a bit actually, you know, SAP, but also Microsoft, who are trying to kind of, like, put the process mining technology already as a service into their product. So, you know, you would just buy SAP, no matter what kind of a system would it be, but there would already be integrated into that. Is this also something where you think that the future of the field is kind of going into? Well, if I could really foresee the future, then I'll probably be working anymore. If you ask me, if it's a likely future, in some aspect, I would definitely say, yes, that it could be there because basically what you are talking about is a certain, let's say, commuterization, right, of technology. I used to pay a lot of money for Excel, right? Now, Excel is even online and it's actually even having a lot of options now, right? So, these things do change, as we have seen also in the past, and why shouldn't they in the future probably even accelerate it? So, that is clear. It's the same thing, you know, like if you come from a Unix world, you have to have a piping for a copy paste, right? If you're in a Windows environment, copy paste is for free. It's in every application. So, certain things are there, probably. So, to what extent I don't know, but I think the discussion anyway would be probably more about the outcomes, and that will then relate into technology. So, let's talk about multi-event block. Let's talk about traversal process flows. Let's talk about dynamic case key definitions. The outcomes are better doing that than fine. But let me tell you, it's just very short story, you know, like my father was also a software engineer all this night, and there was really a colleague, really fresh out of school, and he was the best of the best. We were talking like 20 years ago, so Java was new and stuff. Problem is, no one understood this code. He could refactor 20 pages down to three lines. I'm not kidding you, but then nobody understood it. And then he would actually tell the people, "Are you stupid?" Well, that's a problem. I guess there's some technical prowess, but that also needs to be some social skills as well. And I don't know anything where you can really be one man's island forever. For some part, because you are so at the edge, I believe that, but I think it's much more important to bring the masses forward. And therefore, I think really the focus should be on how can we deliver value fast, and how can we make the starting point with less barriers to entry here? Yeah, exactly. And this is something that I think the complexity part plays into. Yes, we can do multi-event log. We can do dynamic case keys and all that. And by the way, you listing all those made my hair stand up, because just imagining all that in one data model is something to cry over. So that's the complexity, right? And even people just looking at one data model with one case key, with one event log already struggling already to really get a grasp of what they're looking at. So just adding in a whole bunch of complexity, I think obfuscates the value that you're trying to get to. I think a lot more than it is really helping you. So in terms of that, you have mentioned that the process mining tool is more of a commodity nowadays. Can you a little bit elaborate what you meant by that? Yeah, actually, I have to be a little bit very specific. I definitely think that the tool of process mining is still going to grow quite a lot. So I think personally, I think the market with this scratching, but it will also be accelerated. What I mentioned with the commodity is the technology, not the tool. So when you talk about the technology, as we know it at the moment, Pro-M from there was there around, I don't know, 25 years. And I don't have a chance to actually use the open source, it's actually pretty cool. And so I downloaded the other day and it's actually pretty powerful. It's not beautiful because it looks differently from a different time, but it's actually quite performant. And I think we need to also understand the technology part and the commercialization and the value generation onto a corporate sense. So therefore, the commoditization, I mean, is in terms of the technology that don't expect any miracles. Like it will create the data model on the fly for you because you're too stupid, but it actually understood what you wanted to do. No, that's not called that path. On the other hand, in terms of application, I could actually really imagine that the tool would be, let's say, allowing you, like when you come from, let's call a course, in course, BI environment, to actually have BI and process mining use cases based on the semantically same data, but the kind of joining in the different kind of context. And therefore, you can actually utilize the ecosystem BI and process mining at the same time with certain limitations, which go less. And then probably my kids will not think of BI or process mining. They will just think of data visualization in all of the context, something like that there. Right. So one of the things that you also mentioned is that your opinions sometimes are a little, you know, splitting people. And I love that actually. One of the things you mentioned is that and we already touched upon this topic a little bit was that this hyper automation is generally a wrong approach when it comes to process mining. And I would love you to get a little more specific and deep into this into this topic as I do see it very often myself. And we talked about it with Patrick in some of our previous episodes as well, where we simply received this request. Like, let's automate this part of the process without having better understanding of what it actually means. And we are always, well, technically doable. We are always trying to kind of, you know, put our feet on the break and say, let's take a step back and let's actually understand why are we automating this? Because if you have a roof and it's raining inside, well, you wouldn't probably buy a pump to suck out of the water of your house, you would try to probably fix your fix your roof in the first place, right? So what is generally your take on incorporating automation into into process mining? So just to fill the pot of jar with one euro, I have to put then, right? Like, if you automate a bad process, it's still the automated bad process. But actually, I want to go one step further. And Peter Drucker also said, there's nothing so useless as doing efficiently, which should not be done at all. Right? Yes. I think also sorry, Peter Drucker also said strategy is not what you need to do, but it's also what you don't do, right? So I think the topic of automation and starting therefore why this is very, very, very important. But if you want to meet concrete and let's actually talk technical because even on the technical standpoint, many people get it wrong. I believe that actually in terms of the algorithm, the function to automate something usually exists already. But what does not exist is actually the master data and transaction data to make it work in the way you want it to happen. So therefore, I don't actually believe in this, let's say so-called single source of truth. What I want is truth everywhere at the same time and the truth is wrong everywhere because it's better to be wrong everywhere at the same time than being sometimes right and sometimes wrong. Therefore, I don't have the mother load and I do synchronicity all around the world. And to make it very concrete, so you understand what I mean, is think about the movie in Netflix. Is it a different ending than my movie because I watched it in Singapore? No. So same content being delivered on different devices at any time consumed in different ways. So what I'm trying to say, therefore, is if in Sweden there is the next iPhone coming out and they have specifications, I am pretty sure that auto processing can do it. But if the master data is different in one country than the others at the right time at the right place, then the automation, quote-unquote, will not fail, but the outcomes are not desire ones. So therefore, automation actually does not start with the algorithms necessary to do so, but actually with the data at the right place in the right context. Yeah, I always say content is king, context is king-com. And therefore, we need to understand that when we talk about master data, it's not about me saying the iPhone says the master data is 3 kilograms and we know it's wrong. That's the easy part. The harder part is the provisioning. I need that data field at the right time to be available because if I go to the hotel and I go to the breakfast and I see, oh, scrambled eggs. I open the thing, no scrambled eggs. I'm not going to wait for scrambled eggs. That should have saved. I go somewhere else. And even from the kitchen, somebody comes. That's a provisioning part. And the second one is really the topic of ambiguity. I said, oh, nice vegetables. I open it up. I see tomato. Tomato is a vegetable, but actually I didn't see a tomato. I saw tomato soup. And then I go at the other one. It says vegetable soup and I see others, right? So the ambiguity of one thing being able to be many different things is not so hard to solve, but the algorithmic necessity to automate has nothing changed, I think, and how many years now? Probably 80, 90 years when we started with first algorithm. With three statements, you can solve any solvable logical problem, right? It's a loop, it's assignment, it's a decision-priced condition. It's automatable. That's not the problem. The context has also problem. Now, as a head of transformation at Signavio, do you find that a lot of customers or a lot of clients there are very focused on automating things. They go on to this digital journey of, okay, we have now insight, we have data literacy, we have transparency of our processes and all these things now. That's great. Now let's go fix our problems. And the quick wins will can just automate things, right? That's usually kind of a knee-jerk reaction to seeing things that are manual processes. Okay, let's go automate this. Let's go automate that. Do you find that it's more beneficial to kind of take steps back and involve change management to solve these problems in the first place? How do you, as that digital Sherpa in a way for that client to guide them through on how to best approach these automation things and specifically when they're trying to solve these quick wins in the best way? Honestly, I would love and wish to see that, that you know, it's that back in order to actually be faster, right? Sometimes there are no certain shortcuts like that, but I'm also very realistic and pragmatic. So if you really want to change, unfortunately, there's only three ways and it has to hurt really a lot, right? So if you do the wrong thing to hurt quicker, but still manageable maybe that's better. You cannot even imagine in how many situations I told them, hey, if we do that, it's exactly like driving the car with 100 miles an hour against a wall. I tell you after that, you can trash a car. And then this is okay, tea. Then take the car and drive the 30 miles an hour against the wall. Why? I drive the car with 30 miles an hour against a wall. I break some lights and some bumpers. And then the manager told me, luckily, we did not go with 100 miles an hour against the wall. We only got some bumpers. Now we saved money. Whoa. And that is basically the problem of automation. Let's automate. Let's not have the menu work. What do you do when there's a people? Do you have process costs? Do you then engage and involve them in other topics? What do you do with that free time? So to speak. So therefore, again, why do you want to automate? If you want to automate to do then the other things necessary in order to realize your value, that's another discussion to have. And it is really that so many people perceive change management as a very cumbersome thing. And yes, it is not easy. However, I also know the right mindset with the right toolset with the right ecosystem. It's not as hard as you think. And that is the same thing. I mean, you look on the internet, I used to weigh many more kilos than I do today, right? I could not lose all of that in one day. I needed to first start changing my nutrition, my eating behavior. Then I also did more exercise. And then I also did more reading. I did better sleeping, right? And I think this is something what I wanted to say is like it is really a multi-faceted discussion. The last question before we let you go of the beautiful conversation we just had. What's still exciting about this whole process mining topic being there in the walls from the customer side, but also from the creator side now for what could be almost like 10 years in specifically process mining? Honestly, I don't know whether it's actually so much related to process mining itself. And you have to imagine, I was really starting process mining very early and I wouldn't say I lost interest, but let's say I sharpened my focus. So I do understand that process mining is absolutely necessary in most. I mean, I honestly have not seen any business environment use cases where they could not have any value for process mining. It's true. But on the other hand, I have not seen so many business cases where process mining was a solution neither, right? So what I'm trying to say is like, you ask me a red wine or white wine, I can say red wine today, for example, but it doesn't mean I hate white wine, right? So therefore, it is really important to understand I can enjoy both. You ask me what I want to focus on at the moment. And if it has to be a technology topic, I have to respectfully decline. On the other hand, if we have actually a discussion about tea, you talk about people centricism and blah, blah, blah, and you talk about customer experience, how would you then achieve what you want to achieve? And then I would say with a garden shed of tools, meaning I need to have all weapons at hand to have the choice to make the decision autonomously and not given by the vendor. And this is really what I want to say. If I told you guys now that drinking water is extremely important, it's good for your health, it's good for your skin, it's good for your metabolism, I don't think there would be any discussion, right? But don't you think it would be a little bit strange if I told you I just quit SAP SIGNAW and now work for a water bottle company? Then suddenly, that statement earlier one minute ago is a little bit touchy-feely, right? So what I'm trying to say is, if I want to make myself guys happy so that they can have a good order, they only need two things, right? Number one, hopefully a good product. And number two, a customer talking good about it, right? So that's why what I'm trying to say is that when we can achieve the customer compassion for the things what we want to do, that's the best thing. Of course, I still need to evangelize that, you know, it doesn't happen and fall from the sky. But that's basically that's what we do. I can go in and do some fort leadership, but at the same time, it's much more interesting to have a customer say what they achieved in their own journey. Now, I think this is what we need. And also, I noticed that I actually, I would say really actually all customers, all customers have kind of a different, sorry, not different, a differentiated approach on how they want to tackle the journey. And yes, many of them actually need it. Most of them, of course, also go by a process mining, but do I come from modeling to mining, do I come from mining to automation? It doesn't matter as long as they have the ability and opportunity to do so. And that's why, by the way, you also see all the IT vendors are going on the shopping spree, right? So let's see what will be out there in the market, that's right. Personally, very excited. I also love that we went from our talks from T2Y and then back to water. Very nice. Thank you very, very much for joining us in our show. I think I speak for both myself and Patri that it's been a brilliant conversation and very eye-opening in many ways on how you approach things and how you look at things. So thank you for coming and joining us. Thanks for having me. All right. For you, dear listeners, thank you for, oh, actually, last question. I almost forgot to ask, where can people actually go and find out more about you? I'm probably just when they listen to your posts, maybe you just include the hashtag or a link, yeah? So I think the best way to do is actually, which I also believe is the best part is that don't go to TV with TV. I would be happy, but that's not important. One important is go to the SAP Signalway community, which is open to everyone, including our competitors. They want to go there. The community and that is really the important part is really diverse and inclusive. So it's really not about catering only for partners or customers. It's really an open community in the broadest sense. And I think this is a really important and from there, you will find me. So you can type in T with T. So the T the drink first, I don't know name, but from there, I would be around in this domain for some more years. And I'm also very sure to follow your journey, of course, as well. Thank you. So it's exciting. Thank you very much. Means a lot to us. And so you heard it. We will definitely attach a link to our show notes. For everyone, thank you for listening. I hope you enjoyed our next episode reminding our business podcast. As you can see, it's getting really, really interesting. The deeper we are getting into this whole transformational journey. And for me, specifically, it's very, very much fun to really deep or deep diver and so on, deeper, really, really cool. So thank you for listening. You know, if you have any question, you can reach us on LinkedIn. We also have an email so you can send us an email on mining your business podcast gmail.com. If you like our show, rate us on, you know, Apple podcasts, Spotify, or tell about us to your colleagues, to your friends and let us, you know, grow and enjoy this journey together. So thank you very much for listening and I'll talk to you in the next episode reminding our business podcast. Bye bye.