PegaWorld 2026: Breaking Legacy Barriers: Allianz’s Journey to Rapid Modernization

Good afternoon, everyone, and welcome to one of the probably more unusual sessions that you'll attend here at PegaWorld this year. Unusual because we're going to be talking about animals zoo. So but please don't storm out of the room so you're in the right place. We're going to talk about Legacy modernization, about artificial intelligence. So you're in the right place. Um, the subtitle of our talk already gives away the results. So Bert is flying again. So this might not make too much sense to you for now, but bear with us. In a couple of minutes. You will you will understand the story. So before we get started, let me introduce the team to you. To my right is Maximilian Deingruber. He's an executive underwriting integrity manager at Allianz. He is on the business side so he can share his perspective from that side of the aisle.

And we have also Swarna Nayak, who who is the head of automation at Allianz, and she will give us more insights on how the future of the zoo will look like, how automation will be driven forward at Allianz. So. Here's what we're going to cover today. First, we start with the life of the zookeeper. So this means how does the IT executive feel as the zookeeper. And we choose this metaphor because we thought this is quite accurately reflecting what an application landscape of a modern enterprise looks like. It sometimes feels like a zoo. So the zookeeper is really the one who has to deal with all the, let's say, diversity. This is what we're going to reflect on the situation and then dig deeper on old Bert. So Bert, meaning the business intelligence, intelligence, business Integrity review tool, and Max will share a few insights on how the situation was where we took off.

And then we'll dive deeper in the Great Zoo transformation. Meaning what was the team that we used? What was the approach that we've taken? What was the tooling that we've used that we've used? And then we continued to show you the results. So this is this is where we can show you how the new application looks like, because seeing is believing. So we will show you some good insights. And the future of the zoo is something where we particularly want to also share our perspective on how the evolution will continue. So it's not just about modernizing application from old to new. It's really about creating this foundation to develop even further. And then we'll close with some, let's say, very honest lessons learned. And we'll should also have enough time to answer your questions. All right. Sounds good. So maybe then let's do a bit of an activation because we are right after the lunch break.

So quick show of hands. So whose company is operating more than 50 applications in productions in production? OK more than 100. More than 200. Okay, so I see you, I feel you. This is exactly what we were referring to when we talk about the IT zu reality. So decades of new applications, of tender loving care, of quick fixes. And so you end up with a really diverse ecosystem of those beautiful animals. The mainframes, really powerful, but so expensive to feed the beautiful, beautiful custom apps. Quite fragile. And to be honest, most of the time nobody even knows how to deal with them anymore. We have legacy databases who never forget a thing but are so slow to move. And then there are those end of life systems, those dinosaurs that should be extinct, but somehow still around. So this is the reality that we're dealing with.

And the problem is that it gets worse. So every year the zoo gets bigger. New species are being added, new applications are being developed, and a few people know how to really take care of those animals. So sometimes the people have just retired. Sometimes they just won't answer the questions for the thousandth time to this Legacy application anymore. So this is a problem. The visitors are getting distracted and deteriorated, the experience is getting worse, and compliance inspectors are also getting concerned. And this is a place where you don't want to be as an insurance company like Allianz. So the question is not whether to modernize, but how and how fast, because there are a lot of hidden costs associated with this IT landscape with the zoo that you're operating. You have to pay for specialized zookeepers, the people who really know the specifics of the application.

You have to maintain different habitats. So the different infrastructures for your applications, you have to build bespoke security fencing to protect your animals from each other and also for themselves. And you have to manage regulatory compliance and not to mention the negative aspects of the deteriorating user experience. So these are all hidden costs that add up. So as I said, the question is how can we get now to a harmonized modern application landscape and how can we do this fast? So AI is probably one of the levers that we should use. And this is exactly what we did. And Max will give us a bit of insights on the challenge and the situation at Allianz and the targets that we wanted to achieve. Yes, exactly. So I'm here to tell you about a about a real life story and about the life of a zookeeper at an Allianz entity.

So Swarna and I, we work for Allianz Global Corporate and specialty, which is one of many Allianz entities and a zookeeper at at. Our entity has to manage, has to take care of more than 130 applications and those to stick with the zoo analogy, right? Those applications are basically different animals, and most of them live in different enclosures. What comes with that is that you're facing rising costs, because maintaining all these different environments is very, very expensive. And on top, you have very limited agility because adoption to new regulations is obviously very difficult. A big topic is the different feeding schedules and care requirements. You need very specialized expertise for all of your different animals. And a problem that comes with that is that sometimes that expertise, that knowledge is not sufficiently documented. That's obviously a risk. And last but not least, your species. All the species in your zoo.

They don't always interact optimally with each other. Sometimes they don't want to play with each other or they don't even like each other. So you have to keep them in different enclosures. So in this kind of landscape, with this, with these kind of challenges, you're, you're facing some, some bigger issues. First of all, you're in a constant crisis mode because you always have to fix things, fix things. There's always an enclosure that's broken. If you have so many enclosures, there's always something you got to do. Then scaling becomes impossible. You cannot just add enclosures. There is a limited amount of space. And last but not least, I mean the risk potential due to outdated enclosures, outdated systems. That's a big, big topic. Well, it's challenging and it's a challenging environment. But we are a profit driven organization, right? So we still have the goal to achieve a, you know, a cost reduction of 15 to 20%.

So that's that comes on top of all our challenges. But yeah, so much just to, to set the scene. Um, now in this next part of the presentation, I want to tell you about one of our zoo animals. And this zoo animal, as Christian already mentioned, is, is called Bert. And Bert is one of our animals has been very valuable. It's been very a tool that we really love. Um, and it's been very business critical even, but it's also been aging and it's grounded. I would say you could, you could see it as an exotic bird or if you will, even as a, as a dinosaur, to be honest, because it's very special, but it's definitely outdated and it maybe doesn't fit into today's time anymore. But yeah, I mean, leaving the analogy, the zoo analogy, um, analogy aside for a second, I want to tell you what bird is to for you to understand what this is about.

So bird stands for Business Integrity Review Tool, as we already mentioned. And it's a tool that is used by our claims underwriting and risk consulting teams to conduct formal business reviews at Allianz. And those business reviews basically ensure adherence to our processes, to our procedures. And as such, there are basically a governance, a governance control. And the users, the reviewers, they use Bert to, to do their assessment. So they basically they create review sets, they complete structured questionnaires, they enter findings, remediation actions. And then in the end, they create a PDF report that contains the entire assessment. And as such, Bert has been very valuable because compared to what we've had before, Bert, even though it was a very manual Excel file driven process, differed between apartments. So it definitely provided, you know, a certain level of consistency and auditability. And yeah, that's why we appreciate it a lot.

But now why is it, why is it seen now as a Legacy tool? Well, first of all, it's built as an Eclipse Java desktop application, and it was suitable for a rich client and offline capability needs at the time. But it's heavy clients and it's local files. And also the complex updates are now seen as drawbacks. Enhancements are very costly and also very slow due to packaging and rollout requirements and also the the code work. It also has architectural limitations that are a big challenge for us because those architectural limitations restrict quick feature additions, and they turn improvements, even minor improvements into quite comprehensive projects And overall, I would say it's just not designed for the for the speed and the scale that you would expect today. But that's more, I guess, the technical side. Why are our users not happy anymore? There is a big problem for us.

It has collaborative challenges. That's what we call it because our review teams consist out of not just one person. It's it's a team with multiple people. And those reviewers cannot work in parallel in this tool. And that's a big challenge that leads to, obviously, a lot of waiting times and, you know, organizing yourself outside of the tool. So that's obviously not ideal. We have a very outdated, outdated user experience. So you will see that later. I have some screenshots for you, but it has a very excessive clicks. The user interface is not very intuitive in the users opinion. And we at Allianz obviously also have a demand for, you know, modern features like integrations and analytics. Think of, you know, richer reporting capabilities. Think of email and and calendar touchpoints. All of this is not possible in our current tool. And the, the offline file design that I already mentioned, that complicates workflows because it requires you to, to download things, you have to upload it again.

And it's just a lot of reworking time that you have to put in here. And I guess another workflow issue is for us that currently in Bert, it only hosts a small portion of the overall review process. So it's really only the review execution. But there's a lot more that comes to the review. There's the planning, the the file selection. And then afterwards you have the alignment with your stakeholders and you communicated. But right now it only really supports one part of the process, and that's obviously a big topic for us. So consequently, there's just a desire for a browser based live collaboration for our users. And the bottom line is that Bert has been a very valuable tool for us, and it fulfilled its role in the past, that's fair to say, but its high cost of change and all the limitations that I just mentioned made it a Legacy candidate for us.

And with a modern browser based solution, we can achieve all these all these goals, the real time collaboration, faster iterations, and also desired integrations and insights. And with that, I now want to hand over again because that's also obviously where Accenture and then Pega came in for us because with a mix between reverse engineering and forward engineering, we transformed our Bert tool because our business did not want a new birth. They wanted this bird to fly again. So how do you modernize a thing like bird without ripping it apart and losing all the institutional knowledge and the goodness that's in it? So that was the question where we started. And for this we assembled a team. The transformation Ranger team that consisted mainly of three capabilities. So first of all, we had the Gen AI Ranger. So these are basically the people who were responsible for the reverse engineering, making AI understand what Bert really is, which then hand over to the Blueprint architects who are dealing with the reimagination of the processes of the workflows.

And then as a third capability, last but not least, of course, the Pega Platform engineers who will make sure that we at the end have a secure platform that adheres to all security regulations and requirements and can then be deployed to production. What we did from a process perspective. I think where it genuinely gets interesting is how we use GenAI to accelerate the whole process. Because, I mean, two years ago this was just not possible. But now these are actually quite cool features that we can use to accelerate the whole modernization process. So in our case, we basically use two things in a tandem. One is the Accenture reverse engineering Platform. And second Pega Blueprint. So the Accenture Platform was really something that we have developed over the last couple of years. That contains a lot of capabilities and tools to understand the legacy code or the code of a legacy application, and can extract artifacts that can then be used for forward engineering via Pega Blueprint.

So this is exactly what we did. Extract artifacts, upload this to Pega Blueprint, and then have a baseline to reimagine the processes. The big advantage is that we were able to compress this whole analysis process from what could have taken months to really weeks. And one of the things that I think was most important for business, we did not require many SME capacities from business or IT. So AI was doing the heavy lifting. And of course, we would inspect the results and then, uh, take this for the next steps. So that was a really huge advantage because this is what some of the legacy transformations really have as a problem is. There is no SME support available. There is no documentation. So what do you do? You reverse engineer. And um, also the second point that I'd like to make, we're talking about one application now, but the idea is to create really a capability for the enterprise that can be scaled and also executed in other areas.

So something that will take you from the legacy assessment to the The modernization to process transformation and then ultimately towards the Agentic enterprise. Having said this, I mean, this is just an exemplary view of how it could be done. And there are, of course, different modernization journeys. So they might require different tooling. And if you want to know more, I can invite you also to the Innovation Hub to meet my colleagues to discuss your specific case in more detail. I mean, one thing that always is tricky is data migration. And I won't go in more detail because this is a topic for itself. So mechanically what we have done is we have extracted 87,000 lines of code from this monolithic Java based application, uploaded this along with some documentation, user manuals to the reverse engineering platform, and extracted artifacts like business process descriptions, personas, data, entity descriptions, even a scoring framework that was not well documented.

That could then be used and uploaded to Pega Blueprint to create the first draft of the, let's say, 1 to 1 kind of translated application. What's important is we did not only want to, let's say, translate it 1 to 1. This was the initial decision to not include too many new requirements, but really have a lean kind of a transformation. But what we saw in the discussions of the, let's say, Blueprint with business, that there is opportunity to also create some additional value. So after this reverse engineering phase where we created also a bunch of user stories, I think it was 230 user stories, 120 test cases that could be used. And going forward, we had an explicit phase. And I want to stress this, which was called test drive together with business to inspect the reverse engineered application built on Pega, because this gives the opportunity to first check, hey, is this really like I want it to be?

And second, hey, is there an opportunity to include maybe super easy some features that I've been waiting for ages and Max mentioned this. This is one of the challenges that it was just not possible to add new features quite quickly. And by the way, pro tip for business case builders, there is stuff that comes out of the box at almost zero cost that adds value. So don't forget about these aspects. Um, the prototype afterwards was built in 2 or 3 weeks or so. And after a time span of about eight weeks, we really, um, had a working prototype, minimum lovable product, how we called it from inception to really, uh, deployable application. However, I want to be fair, in order to deploy an application into a, let's say, complex environment like at Allianz where you have regulations. I mean, this is not a this is an enterprise in a highly regulated environment.

We had to take a few more steps. So this is where we also took a few more weeks to make it really production ready and eventually deployed it to production. So if we were to repeat this today, I think we would be even faster. So that was the first kind of experiment back then. Nowadays, I think there's a lot more experience and we can even accelerate this a bit more. So now I'd like to walk you through a bit on how the new Bert looks like. And Max has a few things, and he's probably very excited about showing you the new aspects. Yes, exactly. All right. I mean, before we, um, before we go into some screenshots, unfortunately, we're not able to show a demo live here because there's too much proprietary Allianz data implemented in the screens. But I have some screenshots for you and those screenshots will definitely make it more tangible and obviously more visual.

What we've achieved here. But before I do that, let me just touch up on like the most important achievements for us and what this really meant for Bert. So by modernizing the eclipse RCP application to a next gen Pega solution, we went from a desktop application with very outdated dependencies and security and compatibility risks to a modern web based application with layered architecture. We had an extremely outdated user interface with a lot of issues, and now we have modern user interface frameworks and responsive design. And that's huge for us. And you will see that will be a difference that you will definitely be able to see on the screenshots. Concurrent access to a case was not possible. That's this collaboration issue that I mentioned before. And now we're using sub cases so that really multiple people can work in parallel on the same business reviews, which obviously also increases our our efficiency a lot.

Um, the implementation of reports was very complex and time consuming in this old bird. And in the new bird Pega application, we have those out of the box process metrics, reports or another out of the box topic. Um, that's just available to us, which is very useful. And, but also our business reports can just be implemented with ease. But yeah, let me, let me go to the screenshots. So yeah, don't worry. That's not the new application. That's the legacy tool. Um, and yeah, I guess what I just want to show here is it's at the end of the day, it's like an Excel file, right? I mean, you have your tabs and then you import your, your information, your data. There is no like no workflow that supported by this tool whatsoever. It's really just data entry. So we went from this to something that looks like this.

What I want to show here is that's basically the landing page when you log into the tool. And what I want to show here is that you just, you have tools for the users that really also help them to organize their work. They have this, they see all of their case files. They have dashboards that we can, we can configure according to their needs. And we have this, this work list. So it really users work on, on multiple reviews. And this work list helps them to prioritize their work, return to where they left off, return to where they are open to do straight from their work list. So it's really also a workflow orchestration tool for our, for our users. This is what it looked like in the old tool when you wanted to create a business review. Again, this is to show you that, you know, the user experience improvement or the UI improvement.

This is what it looks in the new tool. So I guess it's fair to say it looks a lot better. In the old tool, when you then reach just the the the tab where you actually wanted to put in your assessment, where you wanted to evaluate your files and really input your data. You can see here, it's like, first of all, not intuitive at all. Like it doesn't, I guess when you look at it, you wouldn't know where to click. And it also required very excessive clicking to really get down to the point where you can actually then input the data as you needed it. And this is what it looks like in the new tool. And what I love here, I mean, it's just, you know, the Pega style where you can then see the workflow stage that you're in, the users know what they have to do next.

You have your information tabs on the left hand side. So while you're inputting your assessment, you have that kind of information windows on the lower side that guide you through the process where you and you can, you know, take information out of there. So it's just a huge improvement for us. And yeah, last but not least, that's if you compare that to the to the screenshot of the old bird that I just showed, those show kind of similar workflow steps. I mean, this is where you then where you're in the questionnaire and you fill out your assessment. So yeah, I think looks a lot better for us, for our users. It's, it's been a huge achievement and a huge improvement. Okay. So automatically generated application, basically quite cool from from Legacy code. So to sum it up, I mean there are plenty of benefits for both the animals, the visitors, the zookeeper.

So you're now dealing with a modern maintainable application. Processes are integrated. Um workflows are automated. Quite nice. Users love it because it saves time. And while 5 to 15 minutes per case doesn't sound like much, imagine this multiplied by the magnitude of cases that have to be processed. Have to be processed per year. This is quite substantial. And also the since this is not an expert system anymore, and workflows are much more intuitive. So you don't rely so much on SME support anymore. And eventually it cost reduction goals were achieved. And the security and compliance posture was improved. So all targets achieved. So ROI across all the dimensions. And we mentioned operational efficiency and cost reduction risk reduction. However I want to point out one thing and this is the maturity point at the very top. The AI readiness because now you've created a platform that's enabling AI.

And that can be then even used further. So the foundation has been created where there's a bright future, and invite Savannah now to give some color to the future of the zoo. Thank you. Um, so hi everyone. My name is Swarna Nayak. I lead the intelligent automation and AI enabled productivity for Allianz. At Allianz, platforms like Pega sit at the core of how our business operates. They manage workflows, they orchestrate the business processes, and they help our business with key decisions. Very deterministic in nature. We got them already a few years ago, so when we started looking at AI, we didn't want to add something on top as a separate layer or an isolated solution, because that creates even more fragmentation and complexity in our already fragmented landscape that Christian already showed you. All right. So we are AI strategy has been to add AI to our core platform, such as Pega, because the core platform is where the processes are live, and that's where the real business impact would be.

So our strategy has been how do we integrate AI to our core platforms? So in this initiative, this was for one of our claims initiative. Um, although here it does look like I think the lower part you're all familiar with, it's the Pega architecture, not get into that. Our core claims platform is on Pega. It's been operational and running well for years now. So although here it looks like, you know, we just plugged in AI on top. It was that this morning when Ravi Kumar was speaking from Cognizant and explaining that difference between, um, a production ready AI system and the AI capability, he said that difference is so large here we are trying to reduce that gap. Right. And we, we were also trying to bring in that context engineering that he is speaking about wherein, you know, we add we wanted to make this whole system intelligent by design.

So one thing that we noticed. So we were trying to plug in AI to all our platforms, right? One of them is Pega in this initiative. This one stood out because that integration was so seamless in the Allianz world. We were able to ship this within a couple of hours into production. That rest API that we built between AI and Pega. It went live within hours, right? So what we actually did here is we got the thinking from AI. We have the execution and the workflow from Pega, and we have the governance from Pega. So we brought all these three things together the intelligence, the raw intelligence, and the unpredictability comes from AI. We matched it to the workflow with the human in the lead or the human in the loop in Pega, who will verify the data that the AI has produced. He will look at the confidence score of AI and say, okay, this is right, this is wrong, right.

And once he, he or she does that, they then pass it through through the system of records where the data is stored. So we built in three components. The thinking from AI, the execution from Pega, and the governance from Pega. We plugged it all in together into a single model. So that initiative stood out in that sense because it was seamless integration, and the governance and the execution came in from an already existing, um, system, which was on Pega. Um, so it was not for us about introducing fancy technologies. It was not for us about adding AI into our landscape. Right? It was not just ticking the boxes, it was for us. How do we become more intelligent by design? How do we make add business value? Right. And by doing this, one of the key advantages was we just did not optimize individual processes.

We improved how the entire system performs. So what was the key? What did it mean? Right. What did it mean to our business users? What did it mean to our business? Because this this is a production ready system. This is not a pilot. Right. So this is live as we speak across, um Allianz as an entity. It's across 75 countries. Right? Eight lines of business. Every line of business is different from one another. So what, what does it mean for our users? First thing is simplicity, right? We did not disrupt any of their processes. As I mentioned before, Pega has been in our world for years now. So for them, it meant like they already know what those systems were. They were used to it, but it's all working the way it's supposed to be. But it has become intelligent in design. So somehow AI has become invisible in Pega.

The second thing for the users meant user experience. Initially, they had to go traverse through multiple documents, multiple emails, multiple systems. We have funneled through all all these requests into a straight through processing in simple cases and a human in the lead or human in the loop in complex processes. So the user still comes back to Pega, but everything magic is happening in the background. The third, and the most important part, which our CEO would love to hear is the efficiency, right? So the AI and Pega is doing all the boring tasks like reading the email, summarizing the documents, performing the data entry, and then the user can focus on higher value adding tasks such as claim handling and underwriting. What does it mean for our business? For our business, it meant faster processing time. The claims that would take days to process is now taking only hours.

It means that we have better data quality at source. It means that all the decisions are standardized and not up to the individual or user interpretation. And it also means that the data is consistent. From a scalability perspective, because we connected AI to to our core system. It was. Which was already scaled across multiple lumps and regions. The scalability became very easy. And as we speak now, we are scaling this from Europe to North America. So it has not been, you know, such a challenge. So this has been um, so in a nutshell, like I said, AI became quite invisible, embedded in our system guiding and helping, uh, our users in every business decision. Last but not the least, we also wanted to talk a little bit about what did we learn from this journey, right? The first thing is, as Allianz is so fragmented and nuanced in terms of their IT landscape, we always start with a proof of concept and a minimum viable product that is actually productionized.

It's just not an MVP for for the namesake, we productionize our MVP to see whether. Does it work? Is it worth the investment that we made? We slice and dice to a small lob and region to see whether the concepts that we have in mind with an AI strategy on top, right, so that it can then be scaled in case the MVP is successful. I'm sure Christian and Max have some other lessons learned from their initiative, so I'll pass it on to them to share a few more things. Yeah, I think what I really want to point out here is the this last point on the right side of the slide, I mean, we had this like super fast paced project going on and it was very exciting. And we were very focused on like, what can be delivered by the provider. But we were actually running into this issue where our internal governance processes, all of our internal stakeholder alignment could almost not keep up with the speed of this transformation journey.

So, you know, as it is at a large corporation like Allianz, there is quite a bit of stakeholder involvement and alignments that you need to find. So that was quite a challenge for us. And maybe one more point. It's not even mentioned on the slide, but it came up on the in one of our conversations yesterday. It's also to get the users to buy in on this new tool, because as good as it looks, a lot of the users, if they have been using this old tool for ten years, they don't care how good the new tool is because they know the old tool. And even though it was a not a good solution at all, they were used to it. So they were, you know, they were comfortable with it. So it's it was crucial for us to involve them early, to tell them the benefits, to also explain to them what the change means.

Because having this old Java version, um, it allowed them a lot of flexibility in the way how they use it. And now we have more like a workflow that's orchestrated by the tool, which also comes with limitations, right? Because you have to follow the workflow. So that's another learning from me personally for the, for the business side, I would say where you have to and use your, your users early on in the process and really explain the benefits to them. And a bit of a downer. There is no free lunch. So even though back in the day, everybody was expecting AI to do this all automatically and do the dishes on the side, this is not how it works. So this is actually work involved. You need competent people and you should not underestimate what still needs to be done to make application production ready.

And also the consistency of the team, the continuity. That is something that helps with a lot of expertise. I mean, we were on the safe side with somebody like Oleg from Allianz Technology who knows his stuff. So but this is crucial in a world where automation plays such a big role. You need to have really senior expertise to challenge this, to make sure that it's done in the right way. And I think this is only something that we can recommend to you. All right. So we have talked a lot. Do you have any questions or what kind of questions do you have? And I understand that you need to walk up to the microphone, which probably makes it even worse for you to ask some questions. Oh, excellent. And maybe ask a quick question for that AI integration that you described. What are you using in terms of kind of your AI structures there?

So we are using another SaaS platform there. So we were debating between various platforms that we have. So we finally decided to go with another SaaS platform because they have they were already in Allianz and established on the submission side. Yeah. So that's an enterprise platform. Yeah, it was an enterprise Platform decision then. Yes. But we'll start now looking at Pega Infinity to see, like, how we could maximize the benefits of it. Yeah, we just had a discussion yesterday with the Pega team. Yeah. Thank you. Just a question around. I think it was Christian you mentioned about repeatability and reuse. Um, and the fact that, you know, Burt was the first example of this and that if you do it again in the future, it'll be easier because you've learned so much. So in the context of Allianz, have you taken steps to, let's say, you know, identify additional candidates for this across the business?

Yeah, that's that's exactly, exactly the case. So, uh, and there is even different technologies involved. So what we are currently looking at is Lotus Notes as an example. And I'm sure some of you guys are also still dealing with notes applications. So this is one of the next steps. And I mean Allianz is just one example. I can also tell from experience from other clients. The beauty is that the originating technology is not the limiting factor, because nowadays our tools have advanced so much that it really doesn't make any difference if it's Java, if it's Lotus Notes app, is it is it COBOL? So all those things can be used for an automated modernization exercise. Hi. Thank you for the presentation. Thank you for sharing the insights. Um, I have a question on the last topic which you touched on the user adoption. So, uh, is there anything which you can remember in terms of the Pega limitations, which you had to deal with, especially with some features which users might, uh, just, uh, like after they use like the old system many years.

So yeah. Do you have some insights on that? Yeah. I guess what's, what's tricky is, again, what I mentioned that before, it was basically they could, in the old system, go wherever they wanted in the process, input their data. Now you have a workflow you need to do one step after the other, right? So that's for some users. That's a challenge, right? That's not immediately a benefit from this transition. So that's one thing. And I mean one other thing that we've experienced is that is that it was not so easy to implement ways to go back in the process. So when you like, work yourself through the workflow to like implement functionalities, how you can go back, manipulate the data that you entered in previous stages. So that was quite complex to get this implemented, implemented accordingly. But we got it figured out in the end.

It's just not so straightforward. Okay. Thank you. I have one more question if I can. Um, so can you describe maybe in a bit more details, how was the process of, uh, requirement documents preparation looking like? So I like the part with, uh, like, I think you mentioned some tool from Accenture, which you used. So, but a part of that, was there any process of preparing the documentation about the requirements, which was not part of the Legacy system, but which you still wanted to introduce in the in the new system? I think there were a couple. Right. So, um, the extraction part is actually quite cool because it generates documents that are quite easy to read. I mean, you have basically documents that kind of list in prose, in plain language, what the requirements are. However, during this third stage that I was explaining where we had a sit down with the business, this two day workshop to examine, does the Blueprint look like how I expected?

Is there a potential to include a few more things? This is where we also introduced a few more ideas that then made it eventually to the production stage. Okay. Thank you. All right. I think we're on time. Perfect. So, uh. Well, what can I say? Thank you for the conversation and the many questions. It was quite fun talking to you. Um, maybe the one message that we have is nowadays, Legacy transformation can be done quite easily. And you have had probably a lot of learnings throughout the last two days. And what we can try to show to you is that it actually works. It was fun and it was quite a concise exercise, and we're looking forward to the next steps in this journey. This is Swana. This is Max. I'm Christian. We're around for some more questions and thank you for your attention. Have fun.

Thank you.