What is Process Mining? Vol. 2

Knock, knock. Who's there? Reminding your business podcast. Who? Reminding your business podcast to show all about Process Mining Data Science and Advanced Analytics. It's episode 50. We have made it to 50. Oh my lord. Yaku, episode 50. How are we feeling? We are feeling amazing. Now, can you believe it? No. Today, what are we talking about? Today, we are harkening back to one of our most popular episodes. What is Process Mining? And giving you dear listeners a refresher course with more insights and wisdom and golden nuggets of knowledge that we have gained from these 50 episodes. Let's get into it. Hello, Process Mining Community. Today, we are recording episode 50. And that's 50 with a big five at the beginning. And it's awesome because I honestly didn't see this happening where we were starting. What do you think, Patrick? No, I remember doing two episodes and then, you know, having a meeting trying to figure out, okay, what are the next two going to be? And by the end of those two, okay. Yeah, that's about it, right? That's all there is to say. And I remember the beginning when we were diving into topics, we listed like four. And I was like, um, cool. So, uh, where do we go from there? Yeah, we were talking about, yeah, it should, you know, on Netflix, you have like those mini series, it'll just be that. It'll just be like a one off we do four episodes and call it quits. Yeah. And then I'm so proud that we really made it to 5th, episode 15. Yeah. It's two years, like a insane amount of time. Just for you listeners, there obviously is some work going into this. Just the recording itself usually takes one to two hours. Even if we do an hour long episode because we always have to sync with Patrick. We have a nice discussions and exchanges just like we have today. We're recording this episode where we even want to go with the topics. And the preparation of the, and the research of every guest that's also something that is really time consuming. But very interesting because, um, you know, you go into the depths, you check what the people are working on, you get exposed to interesting topics, interesting brands, interesting ideas. And then you have the opportunity to ask these people the questions that you really, really, really, um, crave for answers. So it's really interesting. And so I don't want to complain about the amount of work that goes into it because, I mean, at the end, we do it voluntarily. And it's just interesting to see how this whole podcast is evolving. Yes. And also it should be mentioned that a lot of work goes on after we are done recording. Shout out to our editor. Thank you so much for all the work that you do. Yeah. Yeah. And also, um, also just, you know, organizing everything, you know, guests, I mean, you, you do a lot of this emailing and, uh, organizing the meetings and scheduling and all this stuff and cancellations and all this stuff. So, you know, there's a lot of, um, you know, not the, not the nicest work goes into this. But, um, it needs to be done. Speaking about credits, also, credit goes to our marketing guy, those who's helping us tremendously on LinkedIn, especially. And then after all to all the guests that have been part of the show, they took a lot of time from their busy schedules just to come here to the show and talk with us, which I'm very grateful for. So thank you for that. And after all, I can't forget about you, dear listeners, because ultimately you are the people who make our show possible and who are, um, you know, pushing us to the next level to the, to, to, to record the next episode and to improve and to offer you the content that. We would like to consume and then you will eventually, uh, love to consume as well. Oh, yeah. And I mean, the response to our episodes has, have just been great. I mean, every time we, we get a message on LinkedIn or an email or anything like that, you know, we, we all, we take screenshots and we send it to each other. Did you see this? Did you see this? Um, yeah. So it's greatly appreciated and also your input to our topics and things like that. And also greatly, greatly appreciated because, you know, at the end of the day, we want to talk about things that interest us. But of course, um, things that interest you and because, you know, if we only talk about the things that we want to talk about, you know, might get a very boring. Yeah. Actually last week a Spotify rep, uh, was released. Oh, yeah. Uh, very interesting to look at it from the podcaster perspective because there's this whole suit, uh, for podcasters on Spotify and you get insights on who listens to your shows and where they listen from. And in, uh, 2022, there were people from almost 60 countries. That was surprising. That was very surprising. Yeah, just seeing the list, uh, we, we can get the data on what countries, those are, I think, the top one is, uh, or top three are, um, Germany, uh, there's US. There's also Czech Republic, I think, on a position for. I might just be my mother. True, true, uh, your, your famous mother. Uh, so that was really cool and I, I also noticed that over 100 people had our show as a most listened podcast on Spotify, which again, something that's really worms my heart. Oh, yeah. I mean, seeing the stats like that are still sometimes crazy to me because at some point you get to numbers, you know, when you see, okay, 10 people listen to this episode, okay, 20 people have listened to this episode. And then you start getting into like the thousands and that's when the numbers stop becoming visual, like you can't visualize them. Like, what does a thousand people look like? And it's, it's surprising. It's really, really shocking when you see a thousand people like, my god, they took time out of their day to listen to our episode. That's, you know, really flattering. We better try our best. Um, but also the context that we are not really hiding the numbers. Um, we usually have around 300 people listening to the episode within first seven days of release, which is very cool. And the best episodes, I mean, one episode is a complete outlier. It's actually, um, the episode to what is process mining, which is also one of the reasons we are doing a little remake here today because we were thinking. Oh, wow, episode two, it's, uh, I would not want to listen to it because one of the comments that we are getting is that we. Uh, improved overtime, which you can. Um, this is nice to hear. It's very comforting. Exactly. And therefore we thought, well, let's do a little remake of episode two because hopefully we can do a little better. We have more knowledge. Um, and we can also hopefully speak a bit better. I mean, that is for sure. So, yeah, those are the main reasons it felt like it needed a, you know, a freshen up. You know, re-haul. Yeah, exactly. But this episode itself got almost, that actually over six thousand downloads already, which is crazy. Probably, uh, when you ask Google, hey, Google, what is process mining? Maybe it gives you this answer. Yeah. I'm not sure, but it's still still very cool. However, um, Patrick also a question for you. What, maybe two questions. Actually, what do you like about doing the podcast and what, what is the worst for you? Ooh, um, that's a good question. So the, the best part is, um, the actual recording, because the, when you get into it and the flow, the conversations, you know, the speaking to the guests, you know, they're very, very interesting people that have a lot to say. A lot of those golden nuggets of knowledge that, you know, um, those are the, those are the great things, right. And also the pre and post show that we have, you know, that's, that's, that's always fun. Like the actual time of the recording, that's, that's great. Um, sometimes the times that we do them are, um, yeah. When we schedule them, it's a little early or a little late. I still remember, um, Friday afternoons, um, or particularly bad, um, and also Monday mornings at seven. Also bad, also three AM ones when, when I was in the US, also particularly, um, tough to swallow. So sometimes the scheduling parts are the, where the times are the worst parts. Yeah. One of the, um, following episodes will be, uh, with Marcel Oderosa, who lives in Australia, times, again, quite crazy. So these are some of the challenges that we have to tackle. On the other hand, just thinking about it, how easy it is, you know, it's just about time. Uh, but you can still connect with people from all over the, all over the world and record something, um, you know, useful. So I think it's still amazing. What I think is, um, really interesting is ever since we started doing live recordings with our guests live, um, when, because I think, um, this is airing afterwards, right? Yeah. Yeah. Yeah. So we have recorded two episodes that were live recorded in a studio and we're talking face to face. And I think I can speak for both of us where that experience was so, um, lively, obviously, because it was live in front of each other that the, um, you can tell, I think, and it was a really, really nice experience. And now I also only want to do live recordings all the time. So, uh, yeah. Yeah. Well, the good news is Patrick that I am lining up some guests for a live shows. I don't want to, I don't want to. So I'm not going to, I'm not going to say the names, but I can already say that, uh, because those are episodes that we've recorded that we will have, um, a process improvement manager from a major farmer company in, in a future episode, uh, 51. And we will also have a high neck and a central excellence team, which you've probably already seen on our social media, uh, in episode 52. So those are already confirmed things that we've already recorded and it will be very, very cool. But there are also some other companies that I'm, um, you know, uh, talking with who would like to come on the podcast and we're just fine tuning the outlines and what we would like to talk about because we don't want to give you always the same topics. I write, uh, we want to focus a little bit more on certain things. Um, we will also have a product mining company coming in, which will be very cool, uh, because these are some of the topics that we are being challenged with in our own implementations. And it will be very cool to hear their opinion on that. And then, well, finally, I would really like to bring the CEOs of the major players. I mean, I hope you mean not all in one podcast and have them do get on. Oh, hell yeah, I would be amazing. An old discussion. So why do you think that you're the best and why do you think that the competition sucks? Yeah. Well, I think it would be a nice discussion, but I don't think that in terms of a scheduling thing that will ever, ever happen. No, but we can dream Patrick. We can dream. Of course, of course. Um, dreaming, where we also, uh, when it came to this episode. So we had a multiple ideas on how to grasp this whole concept because we wanted episode 50 to be something special. Um, not sure that I'm not sure whether us talking about what have been been doing for past 50 episodes special. Well, let's hope it is. What we wanted to do. We had a couple of ideas. We wanted to do like a snippet or just a little cuts of our best content from the previous 50 episodes. But then at the end, you know, if you already heard it, you don't want to hear it twice. Why would you? Uh, so that we just pushed away this idea. And then the second idea, what I really got excited about because I was so enthusiastic after visiting a couple of conferences in past past months. I was thinking like, this is amazing. Let's just bring everyone on. So what I did, I just wrote on LinkedIn. Um, process mining community. We want to, we want you to get involved. And we had an question for them that we wanted them to answer on a video. What does process mining mean to you, which is quite quite personal. Um, I don't know, would be difficult in any case to answer this. And unfortunately, the reception was very bad. So as you can imagine, we got exactly one answer, which was a little disappointing. So not everything we do is a success. We have some unfortunate turn turnouts of events as well. And this was one of them. However, since this is someone who already appeared on the podcast before, Vastlet Shopka from Shkoda Auto, we want you to listen to it now because I think it's really really nice. So here we go. Hello, hello, Patrick. For me, process mining means door opener to value hidden in our processes, as well as possibility to explore another competitive advantage. And also process mining is amazing community around the world, which is creating amazing content for everyone. So what did you think? I think it was very, very nice. And I'm glad Vastlet, if you're listening to this episode that you left us this message. Yeah, thank you for participating. And again, I mean, it's a, it's of course asking people to take some time out of their day for something, you know, that isn't really at the top of their priority. This is difficult. So I expected a low response, but Vastlet, thank you for taking the time and sending us that file. Thank you. Yeah, we will probably try this again at episode 100. So just if we make it to episode 100. Well, you said that at episode two, if we make it to episode four. So yeah, true, true. So it's just double or nothing, right? Exactly. Exactly. We'll just keep going. How long did it take us? How long? When do we actually start? How it will soon be two years, Patrick? I think in February, it would be two years. It's crazy. Yeah, that's so much has happened so much time is passing. Oh my God. Yeah. So do you want to get into into the actual topic of today's episode? Okay. Okay. Let's get into it. Oh, boy. This happens back memories from the second episode. Yeah. And again, just to give you a little background on this, I also asked on LinkedIn and community. And the, the, the answer for this was like all over the place. Some people wanted to go more into the explanation, technical explanation, what it actually is. Some people were saying that this is just a sub category of a bigger, a bigger discipline, which is business process management. So the answers were really kind of varied, varied, exactly. And so what we want to do is kind of rethink about what we were saying in episode two. We are actually using the same show notes so that we can see what where were our ideas and thought processes back then and we will try to frame it once again. So what is process mining? So for me, personally process mining is a, is a discipline. It's an analytical discipline for discovering, monitoring and eventually improving business processes, although not only business processes, but any kind of processes by extracting the relevant data. And putting them together. It's probably as simple as that. And what I also want to mention what I didn't know at episode two, but no now that process mining is a part of other disciplines. It's a, you can see process mining as a sub category of a business process management, a huge area, a huge topic that has been around for years and that process mining very nice compliments and brings up other. Views into this discipline and that process mining is also a sub category of data science because ultimately what you're doing is that you're analyzing or interpreting data in a certain way using process mining algorithms, how you're putting it together. And you are getting some kind of a scientific, not really scientific, but really fact based results out of these algorithms. So these are two interesting or very important mentions that they process mining is just not really a standalone discipline, although it almost appears this way for an outsider. But it is really a sub category of other extremely interesting and variant topics. I totally agree with you. This is part of the whole idea of process mining that it's a standalone thing you can just plug and play into your thing and it will just solve all your problems and you don't really need to care about how your business processes are managed in the first place because it just gives you an X ray and you then you can do all this fun stuff. But the key word being the process and the process mining there needs to be some sort of process that you have defined and need to look at that you need to manage and improve or revamp or the life cycle of processes right there needs to be something in place for all of that information that you get from process mining to make sense. It needs to be placed in a context of your entire enterprise and without the business management, you know that that business process management context the process mining can work for sure you can use it as a standalone tool. But I would argue it's not nearly as effective if you were to place it in that context. So yes, it's like a tool in your toolbox of managing your processes efficiently. Yeah, and one of the explanations or one of the ways that software vendors are explaining process mining to wider audience is that they saying that process mining is an X ray of your business and honestly I couldn't agree with this definition more. However, if you and this is just an idea I got right now if you think about X ray in a real life if you're a doctor X ray by itself won't do anything for you you know you can X ray your leg but if you don't have anyone to interpret what you actually seeing on this picture if you have no one to operate the X ray for you you are probably up for a trouble because it's actually quite dangerous to work with an X ray. So you know if you think about it in this context you get a very powerful tool an X ray if you want in process mining tool but if you cannot operate it if you cannot create a system around it that translates the result and takes actions on the result such as such would be you know supplemented by a doctor if you were at the if you had something if you broke your leg then you have just a tool that. You know can do interesting stuff but you don't know what to do it and that's why created this whole ecosystem around process mining and inserting it in where it really can do wonders is crucial for for any success yeah and this is a really good point you bring up so the X ray is and I think the actual X ray and looking at it is a really good analogy because. Any anybody can look at it a bone or like your femur something and see if it's broken okay cool you might not need a doctor to tell you that your leg is broken if you go if you go in with an X ray and you see a broken like you probably already knew that your leg was broken before you went to the doctor right but it's it's the expertise of the doctor and all the knowledge and experience of the doctor that are really helping you interpret the results of your X rays right yeah and in the same and in the same way. And you can then so again it leads back to the the people behind the tool right so this is one of the key things that I think you and I have both learned it's like it always comes back to the people right always in regardless of how fancy your tool is all the technical stuff that you can do and all the stuff it's all fairly worthless if you don't have the right people the this expertise this experience to help you take all this information that you can now see through this X ray and helps you do things with it and effectively do things with it not just not. Okay, broken bone. Let's go. I don't know use duct tape or no, no, it needs to be like something that has been proven true and with this expertise and experience that you have informed decisions about what to do when you see inefficiencies in your process. Patrick, I have a challenge for you when I was posting that we will be recording this episode I got a very interesting request or questions or actually a challenge from Tim Bosman who will be a guest on our show in two episodes who said guys let's try let's try a 30 second pitch elevator pitch to an execute executed to explain what is processing and why he needs to get. Would you be up for a challenge or would you rather go explaining X ray and a hospital and just lose the idea. All right, let me try. Let me try. I didn't prepare. Let's go. Let's go. You're in a clock. Okay, ready and go. Okay. Have you ever thought about what parts of your business are running inefficiently? Have you ever wanted to figure out what one person is doing wrong so you can fight now? No, that's not okay. No, sorry. No, no, no. I tell that's for Coca-Cola. No, no. So essentially it would give you the view of your business and find the parts that are working not as efficiently and find the capital, the value trapped inside your business that are not allocated well so you can reallocate resources more efficiently and get your business running faster than ever. I like that. If I had to put on my pitch. I like that. I mean giving a pitch about technology that some people or you know it took me a while to even understand that myself and giving a 30 second pitch is very difficult in general. So I think that's a good one. Right. Let's hear it. You also want me to do my 30 second pitch. Okay, so process mining is amazing technology that enables your organization to gain super interesting and actionable insights in terms of your processes. You can improve. You can analyze and you can enhance the way that your business is running, make it running smoother and gain tremendous value from parts of your process, which you didn't even know that you can get any value of. That's nice. I like it. Well, you know what I noticed that for both you and I, we didn't really didn't really mention that like we could be speaking about any of the items. We didn't really say what process mining is, right? Yeah, man, this is difficult. Thank you. Thank you. Thank you for this one. So if you're ever asking me for 30 second pitch for an exit, well, this is what you might get. It's a difficult concept to encapsulate well in 30 seconds. I'm sure there's people that can and have managed in 30 seconds to encapsulate all the things that process mining is used for where it belongs in your business and all these things. It's relatively difficult to do in 30 seconds. Speaking of like explanation of process mining, if we drop this, let's say, high level talk about what it can do for you and where to put it, we can also explain a little bit what it means and what it does from technical perspective. Too much into the depth but mentioning what is important and how does it actually work? And this is the part where I think we can get slightly technical and I actually will read a comment from Becca minister who also interacted with my comment on this episode who says that, you know, process mining basis everything on transactions itself, which is very interesting thought and again, something that is worth mentioning. And the visualizations are done solely for the purpose of humans to analyze and see what happens in the different parts of an organization and systems. How would you explain process mining on a technical level then? Well, first of all, we must start with some base and that base is with process mining and a lot of other tools, always data data must come from somewhere, it must originate somewhere, it has some sort of and the data represents an event. Something that happened inside your business, a lot of things happen inside your business, people order stuff, people sell stuff, a lot of things happen. These points of time, these events must be captured somehow. Usually inside an enterprise resource planning instance that they have running in their enterprise that captures all these things that allows them to do material resource planning and all these wonderful things. A lot of companies have this, they have a whole bunch of data that they've just accumulated over the years about all the going on in their business. So once we have a good basis of data, we can take that data and we can find these events in the data and string them together because we're placing them in relevant context. So these would be our transactions then. Exactly. So for example, we create an invoice, that's an event, that's something that happened and then at some point when we sent the invoice, that's another event. We have created it, we have sent it and what happens when you send an invoice? Well, hopefully it gets paid. So at some point, there'll be another event that is the payment of this invoice. So all these events are happening in different parts and then we can use references. So connections between these points, these events, to tie it together so you can see it in one process and you can say, okay, at this point in time, I created it, at this point, I sent it, at this point, it got paid. And then you can look at that, how much time happened between them and not just for one invoice, but for all of them potentially. So technically what you're saying is that you're trying to string together a piece of events that are happening for some unique object in your system that you're trying to track, be it an invoice, be it a sales server, be it a coffee bean, our favorite example from episode two. And you are trying to look backwards, but also forward in time, what was happening for a specific, a specific unique identifier unique object in your process. Oh yeah, absolutely, this is on a percent correct. So it's very interesting because just on the three events that I've mentioned, you know, because this is an obvious question, okay, you make an invoice because that's what businesses do when themselves they make an invoice, okay, what then do you need to do? Well, you need to send it to the person or the company and then at some point it gets paid. That is very short, the accounts receivable process in a way, right? So in that you've already described with just three activities, the basis of an accounts receivable process with just three simple steps. And those are all logical conclusions that you can draw from just saying, okay, what object am I looking at in this case, an invoice that we've created. How does it live inside my business, right? How, how, what's the life cycle of it? How long does it take for me to get paid, right? Yeah. And interesting is that you could do this analogy and you could put these information together, even in a system in SAP or in some other ERP tool, you could literally track your object through the system one by one. That's the point, like seeing the references, you can always somehow read out from the system. Sometimes it's not as straightforward and or it's just fairly difficult to do. But with some major processes that you probably all know, be it purchase to pay or order to cash or accounts payable, you know what purchase order is out in what invoice and you can actually track this in the system. That's fine, but just doing this manually will take you a couple of minutes, at least if you're skilled as a user, maybe a little longer if you if you are seeing a say before different, well, if you're seeing a say before the first time, maybe it would never get this information out of it. Yeah, maybe I would also point out the visualization part of this comment that is done solely for the purpose of humans to analyze. Yeah, that is also true. I mean, we can only put it in context because again, we are speaking to the doctor. The doctor is only seeing the X-ray in a way so that he can interpret it like the actual bone and everything looks different than it does on the X-ray and it stuns solely for the purposes of analyzing the issue or something that you have. So it's all at the end of the day all becomes visualizations for humans to interpret and make decisions not to mention that the doctor probably doesn't know how the X-ray works in the first place or he might have an idea, but it's not like he could go and fix it if it stopped working. That's why you have those different teams that are supporting you with certain knowledge. And as you mentioned, then what you do then is once you string these pieces of information together and you create what we call an event lock where you basically track your identity, you know, your unique object through the systems, see what happened to it, you can bring it all together, not this one sales order, not this one purchase order, but all of them that are in your scope. Which is, again, before processing this was basically impossible to do and you can then observe analyze and just look into your process from absolutely unique perspective, which is as we all know it this spaghetti monster, this process explorer, where you are looking into different occurrences, into different cases, into different parts of your organization, all in one screen, in this, in this amazing process explorer where you can basically see everything. As on this X-ray that we already mentioned a couple of times. And it's and it answers one of those questions out of all my invoices, how many of those are paid late, right, or and then if you have that ratio, let's try and find out where and why that occurs. So it allows you to in a process context, you know, see, okay, if I, if I filter on now all the all my invoices that are paid late, well, it probably and then you can see, there's some activities that are happening more frequently these events that are happening more frequently in this part rather than the other one. So this event could be the cause of some delay and then you can go investigate and then you can talk to all the people that are responsible for putting these events in or that are sent automatically and then you can go ask, well, why are we doing this, it's causing delays. And it asks you to and allows you to ask data driven questions to your business, right. And this data driven insight, I mean, a lot of these things were intuitions that people had about their business, right, we think it's getting paid late because of this reason, but we really don't know. We have an idea, you know, you always have a gut feeling about why things are working that way, but now you can really come with hard evidence because guess what, you have analyzed every single invoice and you know exactly how many get our beginning paid late, how long it actually takes for them to get paid and all this data driven insight that you can bring to the conversation. But it brings me to one of the biggest misconceptions about process mining as a whole, as a tool, as a discipline, and that is that once you implement it, you can just get results and value, you know, straight out of the door, which is not the case. Again, and I think I'm just going to be using this X-ray analogy from now onward because it's amazing. You know, once you get the X-ray, you probably don't even have a trained doctor to interpret the results. It's not as straightforward to just go and look into process, explainer and say, here's the tool, here are the results, you know, poke the people around and get value out of it. It's really not as simple. It takes some practice, it takes certain knowledge, ideally you have to build it up or you know, not ideally, but you just simply have to build it up, it takes a little bit of time. And that's where you are kind of, you know, creating your own doctors in the company who are able to read the results, who are able to step a little back and look at the process as a whole, but also zoom in and look at certain parts of the process. Of the process that they might be familiar with and then can actually translate these insights that they're seeing into some actions or actionable information. Because at the end of the day, process mining is a tool. It can generate a tremendous value and each of these process mining vendors are going in different direction on how they want to do it. Again, interesting to see, but ultimately you are still kind of stuck with your data with your process and without acknowledging that you need to build up the knowledge, you need to build up the mindset about how you look at the data and what you do about it, how you act afterwards, you will be just left with an X ray that has zero meaning to you. Yeah, and this is a good point. The X ray doesn't fix the broken bone, right? It's not just the doctor scans you and you're fixed, right? And the tool does not help you fix your process inefficiencies. And it's such a good point because a lot of the people that are or the companies that have now implemented process mining are needing to approach the way they improve these things in a brand new way, right? Because now they have this data and okay, how do we best go about using all this data and this insight to better and you know, it's a different way of improving your business, right? I mean, depending on the business, of course, some have established business process management and their more business process focused, right? But for a lot of, I would say smaller companies and things, these are really things that kick start a lot of the business process management in a new way. Let's put it that way. Yeah. It's just that the fact that and as Vilvan that asked what have said it, it's this intersection and a bridge between data science, where on one side, you have these algorithms machine learning data mining and whatnot. And the other part of this intersection is the operations management. That's the business process improvement management process automation and workflow management. And process mining as a tool kind of brings these two worlds together, sometimes they struggle, sometimes there is a clash of ideas of approaches of even egos, because you know, if you've been doing something for a very long time, certain way, and suddenly you have this new way, they're like, well, it's, it's just a, it's just a new new toy in the in the in the field. So let's just forget about it. And that's in itself is unique about this technology that it brings these two worlds together. It's also unique in the fire, maybe maybe not unique, but I also think it's worth mentioning that it comes very close in terms of the KPI is to the things you do on a day-to-day basis, right? Because the things that you do on a day-to-day basis are maybe I'm responsible for creating invoices, right? And sending them. And now we're looking at that with like a fine tooth comb and we're going through everything that you've ever done in the business, and we're telling you if it's bad or good, right? So it becomes, it goes very close to the people, the actual work that people do, I mean, especially with you consider task mining, things that actually look at keyboard strokes and the applications you open and close and things like that. It goes very, very close to your actual, you know, almost life that you're, that you're executing at work. And that comes a little bit too close for a lot of people, you know, it can feel like it's a bit invasive in that sense. So I think that's why it's also fairly unique to process mining comparison to other BI tools anyway. So in a nutshell, a process mining, maybe I'll try this 30 second pitch again, process mining is combining a data and factual insights with a business process management. On one side, you are getting a very specific approach that extracts your data from your transactional systems and interprets it and visualize it in a certain way. And on the other hand, you have your business people who are able to drive some value out of these insights because they get unparalleled view on how their processes are executed. And you can enrich it with amazing level of information and details. You already mentioned the task mining because, you know, back in the, I don't want to say back in the days, but still. Still today, task mining, while there are initiatives and you are trying to integrate it into process mining, it's still not as, not as common, at least I haven't seen it that much. But it's definitely going to happen at some point because you want to know what happens between the moment you create an invoice and you clear the invoice. Some, as you mentioned, the clicks on your strokes are your keyboards, because those are also interesting insights which are still not as ingrained into into process mining, but I think we will see it more and more often in the future. And you are really trying to create a digital twin of your process of your business process where you know exactly what happens, why it happens, when it happens, and how it happens. And that's a unique thing to have in your hands if you really want to steer your business forward. And I think also this is a good point to mention. I also wanted to mention this earlier for the basis of the data, the data that we feed into the process mining algorithm, all the insights that are based on it are all based on one assumption that the data is accurate. The data represents reality, because if it's like 75% of the reality, then your insights are only based on it's 75% of your insights that you could make, or even there's information missing which could impact your analysis. It can only really analyze and really interpret well the information and the data that represents reality. And all the things that are happening outside of the data, you know, all the things that aren't tracked in systems and things like that. Well, that's just gone in the wind, right? You can't see things that you don't have, right? So it's the there's gaps there where you know the information just isn't there and cross the money cannot help you find it. I mean, you can have a guess, right? You can venture that probably something happened here, but it won't really tell you what because the data isn't there. Yeah, and what I found also fascinating is where this whole discipline is going and what kind of opportunities it's opening up because the main focus is still like you want to identify efficiencies. You want to reveal the root causes eventually quantify the impact because if you're doing some if you're trying to do some enhancements, some improvements on your process, you want to know by how much you improve. And again, I would love to refer to an episode with John Mark, who actually basis all of his implementations on this this value framework, he doesn't go into implementing a thing before he and his team actually quantifies what can they get out of it. And then you have these insights which help you understand on how to fix these problems, but then again, knowing where the problems are is pointless without the plan on how to fix them. And that's what I already mentioned that it's super interesting to see how different process money vendors, but also organizations are then inserting process money into their business process management toolkit because you know, if you're going into organization where there's a large enterprise, they there's a high likelihood high chance that they already have. BI tool supplies RPAs other probably they are even testing multiple process mining vendors who knows and the key to this success is to have the overall picture on what you want to use these tools for because I think we've all seen it before. When you give different tools to different teams, they might regress to using the tools for the very same things that the other teams is using it for so eventually you will just end up having two tools showing you two same reports, but with different numbers, which is also ridiculous and then you're like, I don't know which number is correct and I don't even know why. Exactly and that makes you question everything right how can the same things tell me different things right and then well if one is wrong are both wrong you know what's what's the right answer right. So I and I think it also and you bring up a good point the actual work goes in afterwards right after you've implement you've identified your inefficiencies that's when the actual work comes in that's when I would argue even the hard work. Goes into because you need to now establish what where you want to be instead that and also changes hard right changes difficult and in order to change we've done some episodes on on how to change and why changes hard and things like that. And it really showed that the change part is yeah the make it or break it yeah and so while you can out down to the get an answer for why questions in process money tool which it's amazing source of these truths for. You still need to focus on the part that comes after to addressing on how you fix it so why is something wrong well you will be able to get to the root cause very likely or at least identify couple of candidates on why things are not running as well as you would probably like them to. But then you go into this part okay so what do we do about it and then you just open a Pandora's box of ideas and every organization will go in the different direction and I always say that like you can. You can start with just educating the users on how to use the systems in a right way you can. Go into the master data management which is something where a lot of problems are hidden because if you're if you're pulling wrong master data and then change them every time you execute a transaction well that's probably a very pointless and very inefficient to do so this is another approach you can do and what I see more and more is and what I also trying to spark among my own customers is. That I'm trying to to tell them talk to the business you know you have your insights but ask them about their problems what is the thing that is bothering them because. If you if something bothers you in SAP there is a it's going to be very hard to change just because it's a it's a huge ecosystem and making a change for thousands of people is difficult. But these tools that are coming now in place are still young there's still a lot of opportunities in them hidden and they can help you resolve these problems just by asking the question what is the problem they will give you couple of ideas and then you go into okay let's first analyze it if it's really a problem what kind of impact it it holds because you already have the data you already can see the impact and then you go into solution mode and that's what I was saying you can. So that's what I was saying you can start with the education you can fix your box but you can also and some of the tools are offering it go into very much action mode with with what you already have by triggering certain events be it just you know triggering some bots in the system to do the work for you or just creating this this bridge for the for the people who are using the tool to trigger certain actions to create action boards and so on which I think it's all. I think it's also fascinating to really then pinpoint the small parts of the process and try to enhance them one by one and I and yeah I mean what good is a doctor if they tell you why you have a broken leg and okay okay thanks dog now what I don't know you have a broken leg so identifying is one part the fixing is this the more crucial one because at the end of the day you want your broken leg to be fixed and actually I mean this is. I mean this is where the body analogy kind of breaks down because I think you would want a better version of your leg that doesn't break or or something like that you put an ionic legs or robot legs or something I don't know. But that's that's that's the crux right that's I think what we're getting at is the the solution to the problems and that's I mean we're not there yet but I think at some point in the future we'll have some form of. You know solutions to these problems where the process mining can. So just so potential solutions to these problems and not to mention that even this doctor won't give you one recommendation that just solves it all you will probably get a cast on your leg you probably be asked not to do any sports to take it easy on your cast is off but also maybe improve your diet and this is exact same trigger effect that the process mining can cause because. With very high likelihood there will not be one problem or one reason that when you fix everything will just run smoother and then I'm just getting back again into this cycle of improvement because that's a never ending game you will fix something but something else might break or you do some changes in the system that just trigger unforeseen changes and that's when process mining is so useful because it can on continuous basis. Analyze this bench market watch for inefficiencies you know the times that take to execute certain transactions and so on and that's what I mean when when I say that it's really just an x ray and everything around needs a skillful personal to run. And and you can really go down the rabbit hole once you really get into the weeds and start investigating everything in your in your business when you have everything connected you can really. Get lost in in the in the analysis and finding problems have you ever seen that one Malcolm in the middle episode where the father goes and. You know he opens a drawer and a creek so he tried to smooth it out he goes to the garage wants to turn on the light but the light bulb is broken so he goes to for the light bulbs and can find anything and so he then drives to the store and you know and he initially just wanted to fix a drawer but you know you can buy a light bulb right so it's you can really go down the rabbit hole of trying to fix things and that leads you to other problems and it leads you to other problems and things that you can identify. Yeah I haven't seen this episode nor the show but I think it's a very good technology. Should we get to more questions yeah we did get some questions from donish magite from silliness who is asking us actually three things i think we already answered most of them in our little rent about what is process mining. Yeah I think you call it around but the first question is what is the role of process mining in business today i'll take this one it's. The role is that it helps you execute your processes and and see your processes from a high level perspective and helps you understand what is actually happening. I cannot imagine or an organization that would not want to have a clear visibility into what is happening. I would hate it in my own life you know I love data when I always walk around with my step with my watch that is. Gathering every step i take and i love them looking back into history was my average per week and so on. And the same applies for organizations if you are a manager you want to know what is happening and why you don't want to just see. A number KPI this is how we perform but you also want to ask okay can we do better can we do worse why why not and this is what process mining brings on the table if it's used correctly and if you are able to translate the results. Yeah I think I would agree with that a KPI is nothing without a a goal or even an interpretation of is a good is bad and where we try to get to. Right so the second one but trick you want to take it yeah yeah so he asked how does it reveal what dogs are not barking something that be I struggles to do and dogs are not barking is one of those concepts of how do you know what's not happening you know and why it's a little bit. So a little bit you have to twist your head a little bit but. From a process mining perspective and figuring out for example would you like to give an example you can put what this is. You can take this one I didn't like I didn't like it so for example if you have some sort of a learning system where or maybe a ticketing system where you get user tickets because there's problems incident management something like that and you get tickets and all these things and at some point. There's no more tickets so one is either everything's working fine and nobody's complaining and there's no tickets or people can open the ticketing site and actually submit the tickets right it's either one of those. And so that's the dogs are not barking so it's not the fact that you're what it appears to be like you have no more tickets you're doing great well it's just the people can't submit them that's the problem and so. I mean from a technical perspective process mining and other BI tools are in the same boat in this case right if there's no data coming in or something right. You just won't see something right but I think from a process perspective I think it could be a little bit interesting because if you look at activities a B and C and they're usually all within three days and a is followed by B almost always and B is always followed by C. Then you can the event of C is the one that you're looking at for example then knowing that B happened but not C for quite some time can be an alert right usually we have C but all of a sudden a lot of the A to B's aren't leading to C's right so in that type of process context you might get a little bit of an answer but even then it's it's still not the most indicative. You know solution to this problem right right well well set Patrick I try and let's go to the last question yeah I was examples of how process mining has driven through Galaday and helped companies do more with less. We have to be honest to be honest this is not an easy question and I will probably frame it and answer it from the other way process mining is an investment. It takes money it takes effort and it takes a lot of education to make it a success. Driving through Galaday is amazing but you have to build up the knowledge to even be able to leverage it first it will not happen overnight it will not happen probably even from month to month and it takes time to build and so what I'm trying to say is that you first have to invest to be able to save something down the line and therefore you know for me as a as a you know when I see some of the sums that are being thrown around how much implementation costs how much. The tool costs I can't imagine this being a frugal thing to do but I also don't run big enterprises so I don't know how big of the drop these costs are for the company so that's for me I cannot really judge that. But if you if you adopt this mindset of continuous improvement I am 100% sure that you will be able to drive your processes more frugal and you will be able to react on. Global events such as such was coronavirus recently where your whole supply chain just went to you know from from zero to from from 100 to zero in the measure of days and if you already have adopted to this adopted this mindset of how you can execute and run your processes. Then I am 100% sure convinced that you can get more value out of it than a company that doesn't have this in place. Yeah agreed however I would like to challenge the word frugal I think if you were to take a process of me driving to work in my car. And I take the worst route imaginable I take highways in the opposite direction and eventually takes an hour to get there and then somebody says well you know you can just take this road and you'll be there faster. I wouldn't say oh that's such a frugal idea you're not being frugal you're eliminating inefficiencies. If you want to call that being frugal I think that's more of a active part of saving what needs to be saved and not necessarily hey this is an inefficient process let's let's make it more efficient. We're not being frugal which is being smart. So I think yes you can save money in your processes overall but at the end of the day you're just freeing up cash to spend it elsewhere. Yeah so frugality is is a word that through me I have one reading this question that went through me so I hope that answered some of it. But yeah if you can obviously save parts or manual work and all these things and labor costs and everything and run your processes more leanly then yeah of course that that freeze up cash and you can not staff so many people on this and allocate them elsewhere but you're still allocating them elsewhere right so you're not really being frugal you're just allocating resources around really. Patrick don't you live next to the office. Yeah I'm a terrible driver. Yeah I'm just kidding. Cool so those are the questions from Danish thank you for those really appreciate it and I think that's a nice rep of our episode. Patrick I think well I will probably leave the judgment on our listeners whether this was a better or worse than our episode too. I certainly hope that we got a little more inside of this and we don't see the topic as black as white as we used to simply just because we accumulated some experience over the years and see the problems see the challenges but still love thinking and talking about it really. Yeah and I mean all the things that we have mentioned these things that we have incorporated into what is process mining is I think a little bit hard to say when when we started when we stop talking about what is process mining but all of these opinions and things that we have mentioned are parts of conversations that we've had previously from either work experience or from conversations you're on the podcast so go back and listen to those if you feel like you need to get your fix about some of the things that we said. And yeah speaking of of course we will be here again in the next two weeks talking with that will be your cousin or sorry your brother in law my cousin yeah that would be weird but yeah my brother in law exactly for continuous improvement right. And that will be also an interesting topic something that we haven't spoken about before and I'm very excited. Yeah ultimately we don't own the knowledge we don't own the term of process mining and the interpretation can be very very white and we love to facilitate the discussion we have been challenged before we have been proven wrong it happens but i'm just happy that we can bring all those ideas and all those opinions on the show and. You know let you the listeners eventually decide on what you want to adopt and what you actually resonate with and what maybe a little less so we are passionate about the discipline I really really enjoyed that last talking about the pick and i'm just happy that you are on this journey with us. Yeah again thank you dear listeners this has been quite a journey for us both and I don't think we would do this if you guys weren't here to you know listen in and contribute and let us know that you're listening and yeah thank you for all that yeah. Thank you very very much if you have any questions to me or Patrick you can find us on LinkedIn we have a LinkedIn page mining or business also you can write us an email I just check today I have their three and answer the emails so sorry for that it's at mining or business podcast gmail.com reach out to us ask us anything if you have ideas for what we should talk about next let us know and if you have any recommendations on any of the future guests we are always happy to. So thank you very much episode 50 is over and hoping for reaching an episode 100 in next two years so a lot of work to do Patrick here's to the first 50 yeah here's to the first 50 if it wasn't nine in the morning I would say cheers with your. Thank you very much and once again we are happy to have you bye bye.