Session Notes: From AI Hype to Daily Reality – How AI Actually Makes Your PPM Team's Life Easier

Apr 23, 2026
From AI Hype to Daily Reality – How AI Actually Makes Your PPM Team’s Life Easier - Transcript
00:00:00
 
 : Yeah, sorry. All right, then let's move on to the next topic. Why everyone also from our side? Um, we hear a lot about AI and we also heard a lot about it this morning already. Um, so the question that we want to answer today is how we can get from this AI hype to actual virtual reality and also show you how that can. So just to quickly start, my name is Sia Hilla. I'm a product manager at CPlace for the topic CPace AI and I'm here today with my colleague Marcel. Yeah, my name is Marcel also. Hello from my side. Um I'm a senior consultant with data citers, one of the main implementation partners from Clex and I'm also happy to give you some insights today. So where do most uh companies stand at the moment? Um on the left you see an AI tool, AI bot. On the right you see a domain specific tool like a KM tool. Uh but um it also applies to any other domain specific tool.
 
 
00:02:01
 
 : So at the end the worker needs to copy and paste from one tool to the other. So has to go from left to right and back. And at the end it's a lot of manual effort and it's error problem. So every single time there's the risk that someone some user makes an error from copying from one side to the other. Um so the question is does AI really increase the productivity at the moment? Before answering this question let's have a quick look back where we are coming from. So just a couple of years ago um everything was still very separated, very isolated. A lot of even big companies we found also in this industry obviously um were still spreadsheet driven. Um everyone had their small little tool, their small little island and it was already quite a big step forward to on the one hand bring all these smaller islands together into into yeah bigger tools like a like a CIC BPM tool to have collaborative decision making automated workflows etc. But also it was the start of or the foundation for AIC insights and where are we now or basically how the future looks like at the end we don't know how the future looks like and we also heard it in another talk yesterday that from technology perspective we don't know um if we are more going into direction of human in the loop
 
 
00:03:28
 
 : or human in the lead so the technology question is still um open open for discussion I would say but what is important um at least in our point of view that you um have a strategic AI integration so that you don't just store AI to your people and um let's see what happens but really from from the high level have a strategic approach for AI let's have a look little look at the facts and figures um so um a lot of workers actually like 68% of this research show that um workers switch between AI tools um multiple times a day um and also only 12% of the organizations have an strategic approach for AI have a unified strategy for that and yeah last um a worker or user is using sometimes up to up to four or even four plus tools or AI tools on a daily basis and at the end it looks nice on the outside but in the inside like using multiple tools definitely decreas productivity. So I want to ask a question. Um who of you is using more than one AI tool on a daily or weekly basis?
 
 
00:04:44
 
 : Oh, that's quite a few. So at the end technology is there um so this is not the problem. Um but it's the fragmentation. So more tools not it's not more productivity. So our approach is definitely to bring in what needs to be together. So AI into the specific domain specific tool at the end and how we do this we'll show you. Thank you so much. Um so after understanding uh the problem that at least it looked like a lot of you know from your daily work um let's talk about how a potential solution can um so first of all um the main foundation is to really move from a very generic assistant and there's a lot of them out there to a very specialized advisor um and if you have used chatbt co-pilot or any general purpose AI assistant you probably know this issue that you open a new tab you throw the context that you think it needs. You explain the situation or what you want to achieve and what you get in the end is a kind of generic answer that you then have to translate back to the system that you're actually working on.
 
 
00:05:54
 
 : So what you need is a chat that is really specialized and fully integrated in the system. And the advantages that it brings is first of all that it's then not grounded on the internet but it knows your context. So it knows your project data. It knows your portfolio structures, patterns, dependencies, and then you also don't have to feed a context every single time because it already has it. If it lives in your system, it knows the pages you're on, the data you're relying on. It can already use that context and you don't have it this context every single time that you want to generate something. Another advantage is that it also just doesn't give advice but it can actually help us in our PVM solutions to act on our behalf. So if I ask it to plan out certain project for me, plan out some tasks for me that can be done when it's fully integrated in the system work. And uh coming back to the initial problem that it's also not about so I don't have to switch tabs, search through some other um maybe also help centers to get additional information on how to use it.
 
 
00:07:03
 
 : but it's fully integrated in the workflow that I'm working in on a daily basis which also helps us to make sure that it's not one-sizefits-all because in the different roles we have different needs for what the assistant should do for me or how it can support me and if it's integrated in the PPM solution it actually allows us to build domain specific assistance with preconfigured system prompts context that you can give it so it's really targeted to my specific needs that I have on my role. Then of course as a last point we can ensure that it's fully based on the permission system so that it can only access data that we wanted to access or that I myself have permissions to view. So you have full data sovereignity with some additional um advantages such as you can for example connect the LLM that your organization already accepted uh you to use. So um when we sum it up, CHBT is a really good general purpose AI assistant co-pilot as well. But what we need is a really specific AI advisor that supports me directly integrated in my daily workflow and in my system.
 
 
00:08:15
 
 : So let's have a look at what we need for that and basically these are three foundation layers. So the first one and we heard it before in the talk is data that still is most important. need to know the data that we need to give to the AI so that it can actually um answer my requests or act my second part is this integration. We want it to be fully integrated. I don't want to switch between systems and it also allows me to already have a pre-config. And another advantage is that we can attach custom knowledge layers. So we all have handbooks or regulatory guidelines that live somewhere maybe in SharePoint um and usually get forgotten and these are great contexts for context for an AI. So having an fully integrated AI assistant is also allows us to connect those knowledge sources as well. And now I think that was enough about theory at first. Um what we want to do now is walk you through a short story um so that we can have a look at some actual examples.
 
 
00:09:19
 
 : And what we want to do is to show you how to move from the desk to the actual boardroom with an organization that we're looking at um that has a lot of portfolios, a lot of projects. Um it's probably something a lot of you um can relate with. And we have brought three um personas for today. We have Tom who's the project manager. So he's responsible for managing tasks, coordinating teams. We have Sandra, she's a portfolio lead. Um, so she's responsible for reporting, portfolio health. And then we also have Marcus who's also a portfolio lead, but for a different sub portfolio. So what he wants to do is to configure his own portfolio. And with that, let's start with Tom. Um, because Tom just came back from a welld deserved vacation, let's say, of two weeks. And so, you probably all know the situation when you come back, your email uh mailbox is full of emails, updates. You have no idea what happened in your project within the last two weeks.
 
 
00:10:17
 
 : So, it usually takes a lot of time to catch up. Some hours maybe to have an idea of what happened. What AI allows us now is that within our PPM solution, we can have a fully integrated uh AI solution like we can see up here with the little button that quickly quickly allows Tom to simply click on it to generate a new change report. So, what you can do is to put in the date when he left on vacation, let's say two weeks ago, and then what the AI does with a preconfigured system prompt, it now goes through all of my data and all the data changes that happened in those two weeks on my project. And then based on that, it gives me a full report outlining all the critical things, what is delayed, what I should look out for, um, and even some additional information. And it of course keeps me in the loop. There's explanations, there's links, so I can jump into those data changes to see what happened. But the idea really is to get a really fast overview within a very short.
 
 
00:11:18
 
 : And I think this really demonstrates how we can move from hours of manual work to let's say a few minutes in the end because I still should read it and maybe check uh with some of those information. Um, but this all shows the power of integrating AI directly into the system. And in this case, Tom didn't even have to prompt himself. So he didn't need to feed in the data, prompt what he wants to do, but you can actually preconfigure um use cases like this directly. But let's go. Sandra as a portfolio need realizes that she has a stem meeting the next day. So it's Thursday afternoon and she needs a report for tomorrow morning. So she basically has 45 minutes to run through all the portfolio data of all the projects and come up with a new report. Um this is also a perfect use case for AI and PBM solution. Um because you can simply look at her dashboard. You see a typical portfolio dashboard here and in the upper right corner we can see that there's three projects behind schedule.
 
 
00:12:22
 
 : Um Sandra realized yesterday it was only two. So something definitely changed but she's not aware yet what that is. Um and here we have a fully integrated AI chat. So she can simply open that and she can explain the situation that she needs to understand for all the projects that are behind schedule. Um what the main concerns are, what are the things that she should raise in the steering committee meeting the next day. And so what this integrated AI chart does is is it checks all the projects, all the data changes and then based on that gives me a fully structured report and overview of the things that stick out and that Sandra should um flag in the steering committee and this is already quite nice. So in the chat form we have a structured overview of what happened across my entire portfolio. But that's only one thing because what we want is an actual report that I can take to the meeting because this is what takes time transfers transforming this data into an actual report.
 
 
00:13:24
 
 : So what she does now is she simply asks again that she needs an actual PowerPoint report um for this data that we just discussed. And this is something where a fully integrated AI system can also support because it's now creating a new page with that report information. So based on the state uh or the date um that she's in, it creates this new report and we could quickly see the the control phase because it keeps me of course in the loop before creating something asking for my permission. And now we can simply navigate to this newly created report page. We see an executive summary. We see snapshot data of the projects of the risk. And then with just one button click, we can actually download a fully generated PowerPoint presentation that is in line with the template the organization already uses for report meetings. So with two simple prompts, we were able to generate a full report that she can now take and walk into the steering community with. So that's something when you think about it, doing this manually probably an hour on a good day.
 
 
00:14:30
 
 : um checking all the data, talking with the project managers, getting updates, bringing all of this into a predefined format and the companion um or assistant in this case can actually support me through this. And with that, let's have a look at Marcus. So Marcus is also in the portfolio era. He's a portfolio lead for a a specialized sub project and he needs to create a new portfolio from time to time because uh he works with different stakeholders and they have different expectations towards the portfolio. So um now let's create a portfolio for Marcus. So he has this AI bot um where you can um just type in a prompt he wants a portfolio chart uh that shows on the one hand the on the x-axis the phase or a phase section so showing the progress over the development and on the y-axis we want to have u the focus layer so depending on which areas these projects are and you see sometimes takes a minute but uh the chart is generated it also extracts all the m data in um in separate workspaces um reflecting that the color coding and at the end you get this really nice chart.
 
 
00:15:45
 
 : he can start analyzing it and um can also do deep dives there. And he's actually doing a deep dive and he found out that one project, the bubble one, the rigid looks a little odd, a little behind. And then he can simply click on the project and then scan the project detail page. He directly sees okay um the budget doesn't look too good. We already have an overland and you can also see who's responsible with project leader for this project at the end. This is not AI obviously as you can see this is a standard dashboard but um it's integrated into our AI portfolio chart so everything is connected at the end so everything is very smooth for the user he can jump from one place to the other very easily. So to sum it up, so to sum it up um at the end what what we wanted to show you today is that AI tools are good but not as standalone. It's important to have them integrated into your domain specific tools. So your PPI tools at the end and and that's what we can offer you with C place and also with data cycles as collaborator.
 
 
00:17:00
 
 : Um at the end for the user it's way more smooth way more easy increased productivity and at the end one system you have the full context zero friction and definitely less manual errors. So why are we here today? Um on the one hand with Zex we have a very flexible BPM tool. uh we have modular apps um that can be customized to your needs. So it's very flexible at the end um and most importantly we have AI directly integrated in the tool. So not only customized software but really AI having it folded in and then on the other hand like we from data scientists have a lot of domain expertise very um long history also farmer clients and um yeah we're ready to transform your process into AI ready structures and at the end not only the processes also what was mentioned before the people are very important so we really have to train your teams make them confident to use the AI at the end So what's in for you? Um especially the the personal deep dive. Um that's that's an offer from our site.
 
 
00:18:09
 
 : Um you can find us at the booth or we can do it afterwards. We are more like a 30 minute session and yeah it's free of charge. So why not if you have consultants giving you a free advice then why not take it and then obviously like after that would be a clarification of objects and uh then the project implementation. So yeah, please visit us at our booth and um we also want to mention the C place date for unit and October this year. It's um online and also in person. So feel free to register. Thank you. We have a lot of time for questions already here. I think we're right. Yep. Questions 20 minutes. Yeah. And 10 minutes for questions. So already having questions here. I find it quite nice that the system goes through a tool, but it also means that the tool needs to be updated. Definitely. Yeah. But we also have a very good solution.
 
 
00:19:28
 
 : can really stick through the developments yet for the customer. Uh the productive systems have very limited down time. So basically we kind of apply the the process how you update your iPhone. It takes a couple of minutes, but everything is prepared back. Um, regardless of the tool we are using, I think that the most important thing is to update the tool so that we have the right information and then you rely on people to the updates on project managers. Do you think there is a way in the future to have the AI pulling the information from the project managers? Let's say asking if I'm sending reminder to the project manager to to enter the data say hey you forget but maybe more in more um oral way so that they don't have to write but just the AI says hey please give me your updates and I will update for you. Do you think this is something we can go to? I think absolutely. So I think that's exactly what was used to happen that the AI has configured in a way and such a system that it keeps the human constantly in the loop and it's one thing is maybe proactively um sending out reminders to update the data but one thing is to pull data automatically um because I mean by now with the technology of for example you can connect your Microsoft so you have access to outlook all the emails if you tool so it can start pulling data from those as well that are important I think for data updates
 
 
00:21:22
 
 : project management um so I think absolutely right I um I'm seeing a lot of great use cases here for for existing team members but one of our big needs is knowledge management excuse me is knowledge management because the promise of AI data turns into knowledge in the human brain and the promise of AI is supposed to be that its So can do that right and so specifically in PBL I'm wondering in your strategy in your road map our biggest need and maybe this fits into the panel discussion coming next right is to retain the knowledge of the previous project managers but most importantly resurfacing when it's needed for conference for guidance for hey you know you might have planned two months for regulatory approval maybe you need six because half the sites are in this region And we have the data, we have the knowledge that the timeline we need more time. So is there anything in planned around the road map to sort of from a from a knowledge retention? So we actually have a knowledge retention um yeah modules in place.
 
 
00:22:38
 
 : Um so far they they are still not connected with AI but this is definitely on the road map. um at the end it's all about the system itself that that's the formation for everything. So you need to have a um structure a system that um yeah observe the data what we saw there with all the project data. Um but also like I think what what you were more thinking of like getting in ex um external data like for example large contracts that you have as a PDF or something like that to also get that information or knowledge into the system and that I can work with that recommendation future management and maybe just to quickly add to that I think in project management we have a lot of lessons right? Or from past projects, we know what went well, what doesn't didn't well. And that's data that we have, but it's usually really hard to access because if you're looking at 100 projects, that's a lot of data to go thro

... [transcript truncated]