413- Technology Agnostic, Business Obsessed w/Piotr Mlodecki

Mike Kelley: All right. Welcome to another. You've been heard. Today we've got Piotr Mlodecki. the chief transformation officer and has had some very interesting experience and comes to us kind of from the world of finance and business administration. why don't you introduce yourself and tell us a little about you?

Piotr Mlodecki: Hello, Mike. Hello, everyone. Thank you so much for having me. And yes, thank you for that introduction. It is indeed the essence of my story. I'm a businessman and was a businessman before I turned into, a tech leader. And I've always seen technology as the means to an end, to a business outcome, a measurable objective that makes your vision come true. and technology is an empowerment and acceleration platform. So that's how I've always seen that. And it so happens that the recent breakthroughs give us incredible tools that we can make those visions come true.

Mike Kelley: Yeah. a lot of those recent breakthroughs are they've also accelerated everything that we're doing. and I think from what I'm seeing or what I'm hearing, we're, seeing some paradigm shifts also, about the ERP and the CRMs and with the use of the AI tools today, I'm starting to hear whispers of the monolithic E, p, s and CRMs are starting to become a thing of the past because we can build our own very quickly through AI. any thoughts on that from the business side? Because I know business leaders are like, oh, yes, let me do that. I don't want to buy them.

Piotr Mlodecki: I think this is, partially true. It is that the pure product development, needs to really rethink their business paradigms today. And if they find themselves being a business enablers, where feasibility, is the key then, or just the execution of the process, then they may face certain challenge. the bar is going up and, for those products, for those companies to survive in the market, they need to redefine the value added that they give to businesses. And it is about the efficiency or the business framework, not by the feasibility or the automation of the process that indeed we can gradually do by ourselves more and more. Now I still see CRM or ERP as a deep backbone of company operations, especially the ERP, because you will add the compliance element, the, taxation elements, the, regional governance, if you do the international business like we do, so there are some elements there where I think ERPs are well suited and I think they have a lot of, chance right now to build on top of that platform rather than fear of being replaced. But a lot of, even if you think about how ERP, work in the modern age is, they often provide the core of the business while giving you, options to plug in variety of add ons or synchronized business solutions that you configure specifically to what your business is requiring. And those are in danger because they may not be as fit as what your business vision, as what you, have envisioned for your business and for your company. But I do think there is a second side to that story as well. And, even before the AI, revolution is, we established the transformation office and like we discussed, I come from business administration and finance, and I think of the business outcome before I ever think of technology that is going to support the delivery. there was that concept of algorithmic leadership and we could see that, from years back that trend that would encourage business leaders to think in that algorithmic manner. when we spoke about iterative programming languages, where back it became clear that the gap between the business vision and the digital execution is shrinking. And now with AI, it is almost non-existent. I mean, you can bring to life anything that you have envisioned, but, what if there are gaps in that vision? What if there are gaps in the logics of your executive execution or the ownership of the processes? You may then build tools and, digitally or AI accelerated processes that would drive you off the cliff. And I think that this is the era of, hey, what do we say? Check to, business leaders, to those who drive the architecture of how companies operate. That architecture still needs to be impeccable. You can consult your AI to improve it, but it is not yet to be replaced.

Mike Kelley: So much in that Few statements like, okay, so at a core of one of the things that you were saying, or at least what I believe that I understood, the difference between the design of the software and the utilization or the execution of the software, there's always been that push and pull between the way it was designed and the way that I want to use it for my organization to achieve the organizational goals. and how those were always kind of competing or seemed to be competing. And that also enabled places for all of those plug ins and all of the additional softwares to come in so that I could customize the primary system to then have that nuance for the way I'm trying to do our business uniquely, so that I stand out amongst my competitors in doing business. that gap is shrinking drastically with the utilization of AI. But, yeah, we have to lean into the experience of those people that have been doing these things for decades compared to just the system that I can start to describe what I want. and then, but if I forget that key component, that, hyper drive off the edge of the cliff that you were talking about.

Piotr Mlodecki: It is very well said. Mike. And in fact, I liked it when you use that word Competition, but there was a little bit unfair competition in that sense, that it was shielding business leaders from incoherent architecture of the processes, and purely partially because of the technical feasibility and because of a time gap. So if you think about it, okay, I run a company, I envisioned that process and that software to drive it. Now I go shopping, I go through make versus buy decisions, I develop or I buy, it takes a year or two years to do. By the time it's ready, the market has changed. I forgot what my initial design was and I kind of take it all right. That's what it is. And then the people start utilizing this, oh, the system doesn't do it for me, doesn't do that for me. But it is what it is. That's our business reality. So that goes away. Those limitations are drastically reduced. but that protection layer from what as a business owner, you could say, okay, I mean, this wasn't it's just not as I envisioned that I understand it's not as efficient. It doesn't sell that good. It doesn't bring as much as much efficiency as I planned for it. But it's because the programmers haven't done it as I told them to, and it took too much time in. The business have changed. So I think that shroud of protection for the business leaders is also going away. And, that's why you need to be in this constant, perpetual self monitoring and self diagnosis in terms of is my architecture, is my process architecture optimal? are the owners of the processes in my organization think operationally and strategically enough to drive those processes towards the aligned outcome, right? Could that be better? Could that be optimized? You think about this grand architecture and the architecture of the processes that become components to how your company run with vastly limited bottlenecks in terms of feasibility. It's almost the other way around. It's like it's everything is feasible. You can just do it very wrong because now you have the abundance of tools, and you can still go ahead and buy a software that used to do it or develop one in a classic way or develop it in the native way. But what if you made the wrong choice, right? If you live with the consequences of your decisions, and that those consequences come pretty fast, especially that you are competing against the others in the market, doing the same, trying to optimize the way that they transform the organization.

Mike Kelley: So, the teams that are doing this self-evaluation and looking at these things, how are you cultivating the teams and your coworkers to have that evaluation and to recognize that they no longer have that shroud of protection of, oh, well, they did it wrong. And now we're just going to have to use what we have and move forward. how are you adding that into the culture? How do you because there's so many people who don't want to change the culture. And but this is a key component. The culture piece, I think is going to be harder to change than the technology.

Piotr Mlodecki: I think it is you personify the agents and you personify the process. You describe the process, preferably narrowly. So what you do, what I do and what, works for me is to have this. Oh, now I get it moment and it comes when we define the specific task that needs to be done. And so even if we do a very sophisticated processes, we look at the, let's say the agentic correction of certain errors in system integration. So it's a very boring back end thing that would require certain support. Once you define that, you have this type of error correcting agents in this type of error. Now you think, okay, I mean, that's something that has a job description. That's something that has security limits, that something that has its own KPI. And in a way, that the work that it does, it's not that different than the work from the maybe a junior position would do. Now you give that agent a virtual place in little teams organizational chart. And I would specifically say a smaller team, a smaller group, accountability framework unit that has a certain, tasks, to do, certain outcome to deliver. And now you say, I'm going to replace that work with that virtual position and, then you get this. Aha. Okay. So yeah, I do get the concept of automation. Now, what you're giving me is intelligent automation. And that's like I said, is not that different than the junior position that you would hire in the back office to make sure that, something works smoothly, that certain errors do not occur or are quickly, fixed or any sort of a repeatable task that now, can be executed automatically with the help of AI. I think it's a little bit iterative, but the first iteration matters the most is identifying a job that needs to be done and being very clear about that and say in HR processes in performance management, let's say that we're facing a challenge with the objective calibration. So I get it. It's a very tough task to do enterprise wise objective calibration. How do you assure that every leader in your organization effectively cascaded their goals? And then those goals on the lower level are actually passing through the very basic smart, analysis metrics? Obvious, right? And maybe in some very good, well defined organization, there are groups of people in people management who can assure that once you're done with your goals, that's a good time to talk about it, right? Where that's where these processes are finalized that we you get out towards the year with very well defined objectives that align with corporate vision, that align with each other, don't conflict with each other, that are properly cascaded, right? So let's say you want to break down this, you can break down those tasks to two or three agents that do specific work. What an HR analyst or a group of HR analysts would do. Or sometimes cases that are not done at all. I mean, we just set up the goals and we go for it. We come to the performance evaluation at the end of the year. We look at that. That's the first time someone saw it for ten months and say, okay, that should never have been the goal, right? so if you define that role, quite specifically, now you have the power to start assessing that at the very beginning and preventing those misalignments to happen or those inefficient cascading of the goals now. So you can build the processes, like this, the agents like this. That's pretty straightforward. I'm sure that any technical person who's listening right now can very well imagine a three or five ways how they can be executed. is that now you're getting yourself an efficient, calibrated goal setting process with the help of two or three agents that were specifically trained to do that. Now that is, let's say the first iteration when you define the business outcome. Do you define the business optimization that you want to achieve? In this case a goal setting calibration. And, but we didn't get to the training yet. And what we do is that we do not always say that every of that solution needs to be successful. The thing is that you do those iterations, you come to the MVP's very quick. That's another beautiful thing about AI is that you can come with a fully functional production level MVP agent within two or three weeks. And every now and then we allow ourselves to say it didn't really work. I mean, this is not what the agents are built. This is what humans should do, or this is the process that should be eliminated altogether. So it doesn't necessarily need to be a win. But if it is a win, then you come back to it and you ask yourself a question, okay, now I'm going to train you, right. So now what if we're going getting to get this goal setting agent to with a corporate intent built into it or with some cultural aspects built into it. So it's not a generic LLM rationalization right now is what makes a good goal setting or bad goal setting right now. You now you're going beyond that, and you must go beyond that because the generic LLM rationalization is what everybody has at their fingertips. You got to be better. So now it's my business vision, my, intent in driving company culture, competing against someone else's. Both AI powered. Coming back to that previous example, a backend error generating agent has been trained to fix one type of error. If it proved right, if it reduced the workload of this repetitive work, you train it to do another type of error, another type of error, and It's not much different as you would train a. Coming back to that example, a junior employee. Yeah. So you're adding the knowledge and now you're scaling. And I would just finalize it with that iterative, I'd say repeatable cycle. Because now even within that agent capabilities, let's say I trained it to address the five types of errors, it was successful. I went to the sixth one and it's terrible. Right? And all I'm saying is like quickly and cheaply step back from it. Don't do it right. Let people handle that. So this is how the evolution of a genetic enterprise management is going to look like, in my view. I think you do have to hire specialists. and it goes around, data architecture and that luckily, I mean, it's a little shocking how quickly that industry or the good business practices are stabilized. I haven't seen anything like that in the market. So, now as we are in, let's say the year three of this, AI revolution and we're progressing towards defining what the genetic enterprise means, what the AI, co-running a company means, the good practices emerge and now you have this, what we discussed before this, selection of ways that you're going to go, whether successful or not. a couple of things we need to remember is that the ROI does not come from the question answered. It comes from a task executed, the job done. And when so when you are talking about, allowing your team to be equipped with a variety of these, agents that they can, use themselves. it's often stops at, let's say, improving the decision quality or automatizing some tasks that are just within this person's ecosystem or a smaller team ecosystem. So, and it is, don't get me wrong, it is tremendously beneficial, I think. and it is also super important in terms of driving the culture change. you need to let people experiment. You need to let them fail, of course, in a secured way. and this is how that allows everyone to open their mind. And this aha moment. I didn't know it could be automatized. But if you want to scale it at enterprise grade, you need to look a little bit different. you need to build those agents more into the enterprise data layer that has been properly architected for the use of agents and corporate chatbots. And you as an organization need to take over the responsibility for training of those agents and to make sure that they don't hallucinate that they go outside of their security guardrails. This is side by side. This is like almost like a biggest dilemma because you actually do need to take calculated risks and be pretty bold in how you implement those solutions, allowing people and teams to fail, allowing yourselves to fail and withdraw from some solutions. And now you need that, smart business savvy security team who understands when to say, okay, I need you to be careful about that. It. And, again, I would say I would anticipate this to be the bigger problem than it is really, unless you are sitting in some very sensitive industry. and there is a lot of space for innovation, but the power at scale comes with first party agents, that are deployed across the entire organization or across the, your entire CRM. For instance, if you talk about the agent force or something like that, a good example, the first party agents are deployed across the entire processes, entire, commercial team, driving the way that, I want my commercial processes to look like and then iteratively trained the way that I want my agents to be trained. if I'm a sales leader, I think that, some things are necessary to be recognized as the foundation of all the transformation and the data structure and data architecture, data foundation is a key. There is no AI. I mean, I think the truth for ever since digital transformation became a thing. But now it is just strongly amplified. And if you experiment with agents and AI structures, even in your personal life, you would probably notice when it starts hallucinating, when it starts, mixing up data, when, something, you trained it, a couple of months back is no longer factored in for the reason that is beyond your control. Now that's if when you want to, drive the genetic transformation inside the organization, you need to make sure that none of that happens. So the principles are the same. I mean, your master data needs to be impeccable. Maybe you need to go from your classic single source of truth to the mdms, and you need to make sure that, they are really strongly governed. And then the same story is about the data architecture that is built with machine learning in mind with, AI processing in mind with predictive, calculations in mind. And it needs to be intentional on the data level. And that is, I think the key win or lose element. We've learned it very, I mean, sometimes painfully, but eventually to the very good outcome with the business intelligence where we're trying to build business intelligence on various different structures and eventually got to the point of driving the majority of effort on building semantic models, not on building the actual visualizations. I think this is a little bit the same mistake that we are doing with the AI, that we just go straight for the shiny thing, we go right for the outcome because we see how beautifully it generates the bursts of code, right? But the really nice burst of code is not necessarily a good usable software. And this is where the concept of this architecture comes in place. And, then when we want to bring it to the feasible. Okay, so now I can develop on steroids and I can generate those bursts of codes. And now I get the concept how I should architect them into something that is a really good, really solid, high ROI business process or a really good piece of software. And then I will quickly come to that data element where you say, okay, but now I need the data structure, the data architecture that is built intentionally. It's not a library, right? it's a gold layer data that has certain, KPIs and OKRs and business processes and business philosophies pre-built into the model. So then once you get to the execution, whether you talk to your corporate agent, not your ChatGPT, your corporate agent, or you put it to work and you ask it to execute a specific information, or you put it outside to the world and you have it talk to your clients. Now you have certainty that it will not deviate from your processes, did not deviate from your values, it will not start hallucinating or it will not start just simply delivering a poor experience so that data layer is incredibly important. Again, what's overwhelming is that this is not as difficult as it used to be. I mean, I've been in the data structuring and data vaults and data warehouses projects, and they used to take months and years. Yeah. And that's not necessary anymore. I mean, the good practices in data architectures are out there. it brings that element of you have all the tools that are there and then how are you going to use them and sequence them? But if you bypass the master data management layer, the data architecture layer, I don't think you are going to land on the scalable, and high ROI, genetic processes. you may patch the holes here and there, but if you want to go on something that is long term, that will quickly become an obstacle. a big surprise to me how quickly the market has arrived on certain definition of what works and what doesn't. I don't know how many times since the beginning of this week, I heard that someone saying that the AI is the biggest thing that happens since the invention of the internet. I deeply disagree with that statement purely because of the dynamics of the transformation. Internet truly took a generation to implement, so we had a lot of time to adjust. and this thing is going to be a business standard by the end of this year. By the end of this year, we are going to operationally compete in terms of who built a better genetic organization, who processes their, their data better. now how I find it is through, Communities through, congresses talking to business leaders, through my networking, through exchanging information. this hasn't worked, to building a close relationship with the vendors because I don't think that, all the platforms go away. I mean, the number of platforms and, SaaS solutions will drastically go down, at least in the short term. That's what I suppose with the going to be replaced by the homegrown AI solutions, but not all of them. So, so if we're talking about something as big as Salesforce, you need to understand what they think works and what doesn't, right? And you build a business partnership with them based on that question. Okay. I mean, I can start from any of the twenty or thirty, early solutions that agile force bring in. But tell me what really worked for other clients, What they came back to say, yes, this has truly made me money and they will come back and say something that was probably very anticipated. The SDR made me money, okay. I mean, okay, but is it not a spam? Yeah, but if it's an intelligent spam, you may not like it, but it is actually brings a lot of money back home. So these are, things that help you make decisions, that our business outcome oriented, it always and forever will come back to that statement that you need to have a measurable business intent. If you do that, you will know after three months, if your agent is driving you towards the direction or if it's not, you scrub it. And from the more general, aspect, it is actually the, MIT promoted concept of what the digital platform is. This is a definition that I really like, which talks about the digital platform as the library of reusable components. I really like that concept. It takes your mind off any sort of a heavy, monolithic piece of code that it is unchangeable for the sheer weight of it. and you start thinking of the organization as that nimble puzzle board that you can rearrange, in almost any ways. And now with the use of genetic tools, you can you get so much more flexibility in how you can rearrange, the architecture of your system. and as long as you are hyper focused on the business result of each of the process, that you're driving, I think the logics in how you arrange it is going to come to you very quick.

Mike Kelley: Yeah. and, or are you going to fail fast like you've been talking about? And then switch up that logic of how it's how you're reusing those components. That's correct. Wow.

Piotr Mlodecki: You make me realize how rarely I hear the word internet. So, like you said, it became, a thing because people keep repeating that slogan. The biggest thing that happens to us is the internet. But before that, when was the last time you heard the word internet? Right? It is not needed because you are much more outcome focused. is like what we do and what processes do we run? do we sell? Do we communicate? Do we advertise? What do we do? and I think in that sense, I would like the word it to go away as well. I mean, in my team, there is it, there is what we call the core IT operations. And it does some of these classic tasks, but we build the functional teams inside the transformation office in my organization, which are more business partners around certain applications or around certain business solutions, including driving a agentic transformation. Right now, it's something we're giving a identity to a team who does data architecture and engineering solutions. And we're putting a huge task ahead of this team because we say go to other business leaders, understand what they need. Help them see what they don't see and help them rebuild their accountability framework, their team's organizational chart in the way that they work hand in hand, with, identity enablers. So I think the word AI will go away. I mean, it is become a little, washed up word, right? We're getting tired sometimes when we hear that and the whole conversation will be more driven towards what do I want to get done? Right? And, the end of this year is going to bring a certain proven solutions that the early adopters do through AI. Some of them is going to be very obvious, even like even as obvious as customer service, right? Once it becomes really efficient to the point that your, clients don't mind and actually prefer to get their basic questions answered through AI, they get it faster, right? So this is going to be a become market standard, and it will leave the space for where you innovate. When you build your unique operational know how, whether it is how you operate internally as a company, which affects every single one, right? Or you turn the AI generated, outcome into sellable products or sellable service. so this is where we are going to compete for efficiency and then how we use technology a little bit similar to how we competed in terms of how we use the internet. But then we eventually got so focused on the business outcome that we dropped the word.

Mike Kelley: Yeah. interesting. I love the thoughts. I too kind of want to see it go away, but I recognize that skill sets still needed, but that skill set is needed by everyone within the organization. It's no longer, hey, take it to the guys in the closet over there, bring them some pizza and red bull. now it's everybody in the business needs to have certain levels of understanding of these things and how to utilize them, to become more efficient to outperform their competitors.

Piotr Mlodecki: this is, during the last year's, Dreamforce, executive summit, there was a panel, led by one of the founders of Salesforce about the concept of transformation And the discussion was going around setting up a transformation discipline about coming up with the Office of Transformation. It reminds me a little bit of my story and my company story, because when we first started the Office of Transformation, it was really hard to put a definition on it. We knew that we need to manage a immense international business complexity, and it is going to drown us if we just start piling up, workforce to address every single nuance of, the types of businesses that we do. So we knew that it would have to be very much information driven and very much digitized, but it was hard to put a clean definition on it. I don't think it is a dilemma these days. And when you think about your technology teams and your transformation teams. it is very important that the technology is does not take the front row that we become technology agnostic. and we train our transformation teams or if they come from IT team, then we evolve our IT teams to become business partners, to become those, subject matter experts. And that is a task for how, organizational designs are being structured because you need to hire people with certain experience that share that, let's say, agnostic approach towards the technology, but they understand the business challenge that your finance team may have in the sense of fpna or your supply chain team may have in terms of demand planning and execution, or your people team may have in terms of, let's say, the administrative processes that they run, that takes their time away from actually, driving knowledge and improving people's skills and satisfaction at work and the list goes on. so you cannot talk about it. Team won't give you that, right? Your IT guy won't help you rebuild your fpna process. But they do have to, they have to join forces in the business, intent of your finance team and people who will help them execute that digitally and with the help of agents. Yeah.

Mike Kelley: So there's a question that we love to ask people on the show and to because we're just very interested in hearing what the answers are and then to find out how on target they are. I'd like for you to predict something that's going to happen in the next eighteen months that everybody in it or everybody be talking about? In eighteen months that we're ignoring. Today.

Piotr Mlodecki: I think that the number one topic that I expect the market to talk about in eighteen months is what the efficient agentic enterprise is looking like. Okay. So it is no longer going to be if should we use it? How should we use it? it's going to be how to manage the hybrid organizational charts that have both people and agents on it. I think we are going to be talking about the security and accountability gaps. I think that we are going to have a broader definition of the goods and the bads. And we will work out, methods to, assure governance, accountability, and security within the agentic-enterprise. and, I strongly believe that we are going to define the process architecture and the data architecture as the key component for, agenetic processes to become successful.

Mike Kelley: Interesting. Because I continue to hear and we spent time on it in this discussion towards the beginning about, the data architecture, but not the process architecture. and it makes sense to me as I've looked at a couple of different agents and building of agents, one of the things that's happening in the background is a lot of process, mapping, that's happening programmatically, and most of us aren't seeing that necessarily, but it's happening.

Piotr Mlodecki: So we talked. I think there are two angles. So at first that we talked about it was, let's say having basically imperfect process, the inefficient process, right? and that obviously is a problematic, although in a organization that runs at a slower pace with more, legacy systems, with some, governance and controls, checks and balances that counterbalance the errors. It sounds somewhat goes on. but I think if we want to drive AI processes and genetic processes, this would come out and we need to make sure that we fix the process architecture. But now there is another caveat that I see as a key factor for that being successful is to build the processes that are native from the beginning. So it's not an increment. And this is a little bit more disruptive because you have to take into account the possibility of scrapping your process altogether and redeveloping it as if, you, as if you didn't know any different, but to do it around AI, and that is a bigger question. I mean, CRM is one of these, I mean, I would think of the CRM process in a completely different way. if they are, AI centric or AI native, if you will. so, I see the architecture of a process. Yes, of course you need a very good logical process with clear outcomes, with measurable KPIs. They need to align with your business vision, but they don't necessarily need to be the evolution of your own processes. The thought of all I'm going to do is I'm going to make my old process was more automated with less errors and faster. I don't think that's good enough. You have to ask yourself a question. if I wake up today in the world where AI is taken for granted and everybody is using it, will I invent that process in a completely different fashion and build the architecture around that? it's, I'm probably blessed with a team who is very open minded and, shares that vision that, things can always be done better. and, they're not afraid to question the status quo. So this is sort of a culture that we try to, reinforce with between ourselves and it becomes less of a dilemma, but, I think that again, it is the, positive side of a very short delivery framework that you have with AI because you can build, proofs of concepts very quickly. so it's reassuring. The risk is lower. I mean, if you wanted to do this before and you had to invest six months and, a significant amount of money just to create a proof of concept, just to say there was a bad idea. I mean, you would think twice before you make that decision. The AI gives us a chance to, experiment. and all you need to do is have the courage to try to basically say, I mean, what, would happen if that process has been there forever? Was no longer there. And my task was to bring it up to speed in two months, right? In a completely new form. and that puts you in an interesting, kind of like a brainstorming exercise is how would I do that? Would I really go back and start to recreate the entire history timeline of how people used to do that? Or will I incorporate all the latest tools and build it in a completely innovative way?

Mike Kelley: Very, very interesting and thought provoking. I got a question for you and hopefully this doesn't come across wrong. But on that team that you were mentioning, do you have anybody that's over the age of thirty?

Piotr Mlodecki: I do, I do.

Mike Kelley: Okay.

Piotr Mlodecki: Beside yourself I do and myself sadly thirty and over and, I don't have that observation. That age is a limitation. As a matter of fact, what I see, people with a lot of experience with data architecture, some of those, of the team members that had a lot of exposure to data architecture and who were tasked with, software development. They embrace this, in a very enthusiastic, very, very fast adopting way because, but it comes back to, and I don't want to repeat that too much, but it usually links to the fact that they are very, solid in terms of what they want to achieve. So if I know what I want to achieve, why not to achieve it faster, better, and with less money or effort. and those guys know it very well. And they are right now on the forefront of driving that revolution. So, I feel like we could go very philosophical about it or we could go over pragmatic about it. And I'll take the pragmatic route. Okay. Is that the big part of the question is what I asked my project management to answer because the business concepts are not, handed over an email. It is a change request. It is a project request. And you need to intentionally build that process of initiating innovation in the company in a way that it aligns with the business outcome and it aligns with the business strategy. And, there are methods to do this, right? sometimes, I mean, if you ask question, what do you need and how urgent it is? And everything is super urgent and everything is justified. If you ask a question, what happens if we don't do it? That becomes a little bit different type of mindset. but, is, about, innovation within a company is a process in itself. So it's not just whoever shouts louder or sends more emails. we do this through project management office, and, we do this through that intent that whatever we bring in and especially if there are conflicting resources or a shortage of resources. And then we make sure that we do our best to objectively evaluate it towards the certain, revenue generation, cost reduction, meaningful, optimization or efficiency improvement. And we quantify it so that we can compare it side by side, look at this and then bring it to the broader panel of leaders who have the global visibility or a cross-functional visibility and say, what do you think? How do we drive the company further? Which of these, should we, pursue.

Mike Kelley: Right back to best practices? I could steal many more hours of your time and just drink. In your experience, I appreciate all of the time that you've given us and, what you've brought to, you've been heard. we truly, I hope everybody else hears and, appreciates what I heard today. And thank you very much for your time. is there any parting thoughts or anything, any parting lessons that you'd love to share and, to help those that are following behind us?

Piotr Mlodecki: there is, actually a, quote I would steal from, Jason Huang. It comes from, the conversation between Jason Huang and, Marc Benioff. But I would, repeat that when I heard from Jason Huang. asked if like, how do you feel being at the forefront of that disruptive moment in history? He said, I never know if I'm running for food or running away from being food. And I think it is a very useful, thought for, every leader to, run for food, to, pursue the innovation to pursue, the cultural shift, never be satisfied where we are. Always feel the breath of the competition at your neck. I think it's the most exciting year we had in a very long time. That is going to fundamentally shift how the organization operate and how they're even structured. And, if we do this with enthusiasm and a little bit of the competitive spirit, I think we're going to have a lot of fun.

Mike Kelley: Oh, I don't know about anybody else, but I'm having fun and truly appreciate your time. Thank you very much.

Piotr Mlodecki: Thank you Mike, thank you so much for having me.