Name: Agentic Finance: Moving from AI Experimentation to Quantifiable ROI
Uploaded: 2026-05-07T10:16:56.146Z
Duration: 1 h 1 min 36 s
Description: Agentic Finance: Moving from AI Experimentation to Quantifiable ROI

Transcript for "Agentic Finance: Moving from AI Experimentation to Quantifiable ROI": Hello. Good morning. We're just gonna give it a couple of seconds just to let everybody join, and then we will get started with webinar. Okay. We are at the one minute past, so we will get started. So good morning, and welcome, wherever you might be joining us from today. Lovely to have you all here for this morning's webinar, agentic finance moving from AI experimentation to quantifiable ROI. So this session is part of our looking forward to work day webinar series, and it's designed for finance teams who are thinking, you know, you're doing AI pilots, but you're not quite seeing the measurable value. And the Deloitte, team, you've got Sina here from Deloitte who's joined us to share some key insights from their newest finance trends report, and we're gonna explore some of those key areas and agentic use cases that Workday is building, to, help deliver quantifiable ROI. So before we get started, I just wanna note our product statement. Paul, who's joining us from the Workday team, is gonna be sharing some of our, existing and future capabilities, in the agentic AI space. And then I just have a couple of housekeeping items that I wanna highlight before we proceed. So firstly, the webinar is gonna be recorded. We will share a link to your inbox within forty eight hours. We do use just for for these webinars, we used a platform called GoCast, and we do recommend using Chrome. This provides the best experience. If you do hit any snags or you need some help, please reach out to us via the q q and a functionality just there on the top right hand hand side of your screen. And also with the beside the q and a, you'll also see a link to docs. This will link you out to the Deloitte finance trends report if you want to take a look at that. We will aim to answer as many questions as possible if you wanna pop them into the q and a, and if there's something that we are don't aren't able to take live, we will follow-up with you via email as well. And then also just to note after the webinar, there will be a quick survey. If you can fill that in, we would greatly appreciate it. It just helps us to shape future content for our finance webinars. So with that, my name is Marianne Carr, and I am the product marketing director here at Workday looking after financial management within our OCFO products. I'm gonna be the host for today's webinar, but I'm also delighted to be joined by our speakers from Deloitte, Sina Basing, as well as my colleague, Paul Abboton. Paul, do you wanna do a quick introduction and then. maybe help? Sure. Yeah. Sure. Hi, everyone. So Paul Abboton. I am part of the office of the CFO product strategy team here at Workday. I've been with Workday now nearly ten years, and then prior to that then kinda led various finance teams in, mainly shared services, financial cert services organizations. Sina? Sure. Hi, everyone. Welcome also from my side. It's true to be here today. My name is Sina. I'm based in Berlin. I'm with the Lord since twelve years, and I'm the work the finalist lead for Germany. So helping clients implement not only Workday, but also helping getting a better understanding of the bigger picture, like, whole the architecture, whole data flows so that Workday is well integrated in the whole organization and happy to be here today. Thanks for having Alright. us. Thanks, guys. So before we jump into some of the key findings in the report and what we're building, I just wanna set the scene for a couple minutes. So AI, you know, it's we all know it's reshaping our world faster than we ever imagined. If you think about twelve months ago even, and the evolution in that space, the, you know, entire industry skill sets, you know, career paths are being redefined in real time. And, you know, we see a huge opportunity with AI, but, you know, on the other side, there is also very real risk. Some of it we understand and then honestly some of it we don't because this is very new technology. And this is exactly, you know, what we're hearing from CFOs and from customers. They know AI is a business imperative, and the question isn't whether to act, it's where to start and which investments will actually pay back. And if we look at, you know, AI and its value, you know, large language models are are extraordinary. They've effectively read the entire Internet, but the problem is that they are probabilistic by nature. They reason and they predict based on patterns, and that's, you know, that's what gives them their power, but it's also where the limits show up. Because an LLM on its own knows almost everything about the world, it doesn't know your business. It doesn't know your chart of accounts, your approval chains, your closed calendars, your auditors, just as an example. And for AI agents to do real work, they need more than just a great model. So like any expert, they also need deep context, so an understanding of how work happens within your business. Trusted guardrails, so proven business processes that help to guide those tasks. And then also, government controls, so the permissions and audit trails to ensure absolute compliance when AI is used. And we're gonna talk a little bit about what Workday is doing, to support this, towards the end of the presentation. But just to highlight, you know, when we think about Workday in tandem with some of those AI capabilities that we're building, especially agentic AI, Workday is deterministic by design. AI is probabilistic. Finance cannot be probabilistic, so Gap doesn't accept probably and neither do your auditors. So we've built the platform around four principles that give AI the rails it needs to be safe within that finance environment. So the first is unified context. So for those of you who are using Workday, you know our object data model already understands how every data point in your organization connects. So agents don't have to do the guesswork then. Process guardrail. So our business process framework makes sure workflow stay intact, within the system every single time. So then when an agent acts on top of that, it follows your policies, not its own interpretation of them. And then always on auditing, so every action is captured and explainable. And the last one then is persistent security, and we're gonna talk about security in the framework of agents in just, a little while. But we've got one security model across the system so agents can't, you know, expose that sensitive data. And I just wanna, you know, put simple, the model brings the intelligence, and then Workday brings in the trust as part of our agent to k I capabilities. And when we talk to CFOs about all of this, they do get the potential. They really do. But we recognize that there's a level of skepticism, and a number of studies are showing that early ROI is often negligible. And the Deloitte team, Sina, is gonna talk about that in just a few minutes. So what CFOs are really asking for lands in key areas, and this really aligns into what Sina's gonna cover in the Deloitte report and Paul is also gonna show you in terms of what we're building. So first area is around radical efficiency. So what we're using agentic AI to do is to take human intervention out of 60%, 70% of that finance work that is essentially janitorial. So if you think about the processing, the reconciling of data, the building of reports, and then enabling teams to scale the business without scaling the back office. Intelligence velocity. So stripping latency out of finance. So information flows to the business in something that resembles real time. So the aspiration, we have here is a zero day close. Complete compliance. So widen that lens of control and anomaly detection across 100% of transactions all of the time. And then strategic partnership, so democratizing the data, natural language reporting, and Paul's gonna show you a little bit of what we're doing in this space. So finance moves from explaining the what to really helping to shape the so what. I'm gonna show you what we're doing here, with the financial analysis agent and also in the area of adaptive planning. And that kind of, you know, is this perfect bridge. I'm gonna bring Sina on in just a second to, highlight some of those key trends within the Deloitte report. Because while capability of AI is racing ahead, you know, at a speed of development that we haven't seen in a really, really long time, the value organizations are actually capturing is lagging well behind. So Deloitte's data shows that almost two thirds of finance leaders are actively using AI, but only one in five are seeing measurable ROI. And that gap between what AI can do and what finance is getting from it is what Zena is now going to unpack. So, Zena, I'm gonna hand over to you to bring us through some of those key findings in the report, and we're gonna look at the ROI gap and then also how to overcome them. Sounds great. So I'm pleased to introduce you to our newly published, trends, which we figured out on the finance report. So you see here five trends, whereas all of them are bringing together what we wanna see within today's sessions. First of all, we have understanding the gap and then how to overcome the gap. But let us start, with the trend number one here, the speed priority. So based on the report, 60% of the finance leaders see here that they're really uncertain. So that means when we look back, planning was, like, on a yearly basis or then it becomes on a monthly basis, then it was, like, sometimes once a week. But now living in this uncertain world, we have so many changes so that we really kind of, like, see that within the planning and reporting space, we need to level up here. So we need to be able in terms of from a financial reporting space or from a workforce reporting and planning, space. We need to be able to do ad hoc reporting and planning, so moving and shifting, not only on the financial side and on the workforce side, but also as an organization as a whole. So, for example, do we need to kind of, like, change a little bit our business model? For example, do we wanna go into a center of excellence maybe? What does it mean in terms of our organization? So within this, speed priority, we really see, like, planning scenarios and also reporting. We also often see here, within our clients that in this really fast changing world, there will be also a lot of new stakeholders within the organization. So all of those node stakeholders bring also new ideas. So that means that the report which you might have used for the last ten years might not be sufficient enough. So as an egg hot request, this new person or new stakeholder brings in new ideas and also needs new reports. So really see here in the trend number one, changing, the speed, how you report, how you plan, and also the frequency underneath. The second trend here, we see finance leader are stepping us as strategy leaders. And when I look back within the twelve years with Deloitte, it was really clear. CFO is there to really kind of, like, have an overview of the budget, have an overview of the profits, and also of the costs, and was more, like, seen as it's necessary, like, the finance organization. But now in this kind of work where they're changing so much, we see that, especially in the finance area, CSO was stepping at more as strategy leaders. So, when you look at a typical CV, this was changing. So now 20% of the usual work is now determined, to more strategic thinking. So finance is not only a necessary part of the company anymore. It's really like your business partner because you need to understand where the data is coming from, what the data means, and what we can really do with it. So what does it take, what we can make with all the data available, and how we kind of move into the next next step for the company here. As we see here, the trend number three, focus, precision, and discipline, it's really I think it's kind of a key part of the finance organization. Of course, you need to understand your your cost management and your spendings. But what we have seen in the past really is, like, having a lot of reports, doing manual work, and not really kind of, like, saying, okay. We have all the data available, but what's the next step? So AI can really help here to kind of, like, having the overview, comparing data, and then also picking out, okay, what we wanna focus on. So really getting a better understanding of your cost spendings, but on the other side, of course, also on the profits and the revenue you are generating. So bringing up a big step there, we did taking the data and forming it to move to the next stage with the company here. On the fourth trends, we have the journey to agentic insights. So, of course, what we have seen as AI is on top of everyone now, and we see all that also all the leaders getting, like, goals connected to it. So everyone is okay. Let's let's move forward. We need to implement something with AI. That's great. But on the other hand, it's also first do the basics right. And when we mean do the basics right is getting a better understanding of your architecture, actually where your data is sitting. And then also, okay, do we see here that maybe we need to move from non promise on premise driven to a cloud version? So getting the EIP system in the right state so that the EIP system or the solution provide as a good basis and then putting on top AI or solutions bringing AI enabled already, so that we can really see the shift. And then that agentic, AI is also helping in terms of the organization. So for example, in the finance area, we often see that, data lacks information between reconciliations with all the subledgers and the general ledgers or also from a banking perspective that all the data comes together, is analyzed and bringing in value here. And then also looking at the fifth trends here, including tech talent and finance, we really see a big shift here. And this is also connected to the topic which we see, in the trend, number two, that finance leaders are becoming more strategic leaders, get somehow connected because you're not only the finance organization that brings in, all the figures and then someone else is taking a decision. We really see here that also the team which is working in this area is combining now the finance know how, but also the technology know how. Because we are realizing we have all this great data available. We can use it. We are stepping in into to a more business partner of the whole organization. So it also means that, the accountant, it's not only bringing finance knowledge, but also has a touch of, technology understanding and is interested to use technology to kind of like be the perfect business partner within the organization here. Wrapped up this one, we're gonna move in understanding the gap. So what is actually the output of the finance trends here? So we are seeing here that when we wanna understand the gap, where are we actually right now within the AI transformation? So we see here from the left hand side to the right hand side that, 63% of the finance leaders have fully deployed and actively used AI within their function, I think, which is actually very great. But then we see when we look to the right hand side that only 21% of the final seniors have fully deployed and already see those investments in delivering clear and measurable value that can come to an understanding, right, okay. Maybe we have implemented something like an agent, but actually the data which the agent is using and all the security around, it's not kind of like that what we have anticipated, because the basics are not right. And maybe we have taken two steps four, then kind of, like, previewing the basics. And then when we go to the right hand side, we see that only 14% of those limit clear value have already fully integrated AI agents into specific areas within the finance function. And they are really wanna bring in an example, from our projects and client what we have seen here. So, actually, when I look back, the implementations when we have started are really okay. We're going into a finance, transformation. We implement implementing a new system, and we wanna kind of, like, use all the standard processes and also wanna do, like, all the great security which comes within the new system. So this was kind of like really the main goal actually back then. But now when we look at implementation, it really comes okay. Let's have first, so we have chosen our system, but we are going into have a strategic, conversation with you guys because what's what's important for us, we have so many stakeholders within the organization. What matters are our KPIs, how we wanna steer the organization. And we really wanna use the new ERP system to bringing in value and giving us all the insights here. So and giving us all the insights, it's coming back to kind of like how we can use AI agents to really support us. Because they can support us if we are really like realizing, okay, where do we really wanna use AI agents? And there we have seen that our clients are coming back and saying, I'm struggling in reconciling data. I'm spending a lot of time in preparing for reports or going in the month end closing. There are so many steps that we need to take here. Actually, we are more focused on kind of, like, delivering this in a short time frame rather than looking ahead. So, I think here when we see the gap here in terms of only 14%, the fully integrated AI agents might have come from the lack of understanding, where you kind of, like, really can drive the value. I think they are great example of agents out there, but all we really measure, okay, does it really help us in our situation here? And then coming back, to the middle here that only 21%, of the finance that have fully deployed, all the, AI deployments or, AI specific that they have seen here. When we go into a conversation, leaders go in within their goals. I don't know. My goal is to implement, a new AI solution and having five agents, and then, we are actually, little bit going around within within the first step here. So what I do mean with the first step is they're having great ideas, but also coming back to the basically, what you really wanna reach with those AI solutions and, are you really using it? So for example, we often see that they have great ideas and they also create a road map, but they forget to introduce also the workers' council. So they are, for example, in Germany, certain rules within the workers' council that, you need to consider in order to introduce AI to your organization and how AI can really help you here. So, those are, like, two examples which, we have seen. And considering that what we wanna take away here is that, your basics might put me in the right position, and you might have kind of, like, goals and you have stepped into the goals and wanted to achieve them rather than analyzing, okay, what are really, for example, our manual processes where AI can help us to automize it or can can help us to save time that we can focus as an organization really on on those topics that are that really measure for us. So, therefore, only 63% of the finance leaders have fully deployed AI use cases within their functions. And we'll see. Let's go to the next one. And, also, when we remember of the of the figures we have just have seen, we have identified within the trends, that there are several areas where we're really challenging AI's journey. So the first one is the topic of data security concerns. And we also see here, we have AI leaders, which are flat in green, and then we have early AI journey peers, which are reflected in gray here. So we've seen here, like, we have top challenges. For the first one is this data security. So the ones which have kind of, like, already implemented AI also see, like, 57% that they really have kind of like a concern that they really wanna step up in terms of data security, which also comes back to the fact that some of the organizations struggling with getting approvals in terms of, the workers' council who are also lacking, okay, now does AI really have like all the information within an organization? Is it really what we want? Do we need to change it here? So there are really uncertainties that need to be tackled. But I think it's great that they have already kind of like get into the first place, which they're realizing, okay, we have implemented something, but now we also need to step back a little bit and then think of, okay, how we can really protect our data here in a certain fashion that it's giving us all the information, but only the right person is receiving the right data here. And then also we see that, within the second challenge, it limits expertise. And it's also coming back a little bit of the trend number two and five, which you might remember that the CVs are changing within the organization. So finance leaders are becoming more strategic partners and also that the talents working within the finance organization are coming back and, kind of like changing a little bit their role in terms of, I'm not only the finance expert now, but I'm also driving into more technology, items here to support the organization as a whole. And I think what's really interesting that on the one hand side, we have the AI leaders, which see the challenge here. But on the other side, they are also the early AI journey peers, which also see the challenge. So that's quite even that both of them experiencing the same. And this is also what we see within when we, for example, implement Workday, within our clients. So actually what we have seen here is when we implement something, they are really kind of like persons with us specializing in the finance organization within a certain process. And they have used the system, like, for twenty years and have never changed. So they actually know that what they wanna do next year on the May 7. However, now with introducing a new EIP system, which is also bringing a, AI enablement, we really see that the work of the employees is changing because the system or the solution already brings in so many standards and so many automation that we really need to think of, okay, what's gonna be, the role of you in terms of, okay, you have worked 40% of your time for a specific manual task, which is now automated since we have introduced a new system, or also new, AI solutions. So now, actually, your work is a little bit different than before. And you can also use actually the technology to also identifying what can help you in your daily business here. So that's a great example, that we see here that limited expertise is only within AI, but we also see it when we implement a new system. And then the next one is the regulatory, complexity, which we often see. Especially, when we think of, Europe here, we all have different regulations. We also have a lot of clients which are operating not only in one country, but which are operating in several countries. So that means that implementing a new system, implementing a new AI solution must meet like all the regulatory requirements. And it's for both, like for the AI leaders, and also of course for the early adopters here, which we see because standards are all equal for them. And also of course the governance is also learning when we kind of like develop all the all those solutions and getting, adapted to those, that also new regulations are coming up. So therefore no surprise that it's really similar. However, where we see a gap is, legacy technology. So challenges, which we have seen here that AI leaders, have a little bit of lower of a challenge of the legacy technology. And that might come from that before they have really kind of, like, implemented their AI solution that they reviewed, okay, what's my actually basis? Where does my data is coming from? Do I have the modern system in place, which is actually capable of connecting with the AI functions, whereas we see that, the ones which are in an early stage of the AI journey really have a challenge here compared to the one, which are the AI leaders. And we see this often that we're getting requests from companies, which are kind of like on an order system version, like an on premise version, and now they have, like, AI goals. So, actually, the first conversation that we are having is, okay, where's your data sitting? How does your architecture look alike? How can we help on the short term? But actually what would help you is kind of like on the long term, have a modern architecture in terms of your ERP system here. So therefore, actually no surprise of the trend that we see the gap here. And then what we have seen, of course, and what we are also talking about right now is the unclear of the ROI. We see here that AI leader, so which ones which have been here, have employed something, getting to know the AI solutions, having a little bit of a challenge. Also getting a clear ROI. But what we see, of course, if you're newly deploying an AI solution, you're kind of like, in your journey, like in an early stage, you're not getting right out the ROI, which is understandable, because you are just within the start here. So having those challenges, how do we overcome them? And I would say, let's have a look here. And we have summarized those in three main steps that, the survey is recommending. Like, the first one is investing in the technology infrastructure, and this is also what I tried to kind of, like, inform you about. We have seen that, we have goals, from an II perspective. But if your data is not wrong, if your architecture is kind of, like, really old or you have a very fragmented architecture, it's also really hard from an II perspective to deliver you the value that you wanna that you wanna see and that you also wanna grow within your organization. So first things here kind of like reviewing your infrastructure, your architecture, and also where your data is sitting. I remember two years ago started, with a company. There was a headquarter in Germany implementing Workday, also kind of like rolling out, to to kind of, like, a a global organization. And then we actually start okay. We have, so many kind of, like, we have modernized your architecture. But on the other hand, you have so many legacy systems. So we really need to identify where sits your master data. So which system is your master data, in which kind of state is your master data here. So having this first view, understanding data, architecture, processes is really important here. And also, when we reflect on the trends which we have seen, that a lot of companies were struggling in terms of, the uncertainty, which is coming around, in terms of planning and, not only financial planning, workforce planning, but also organizational planning. This could be also really good, time and place to also reflect your kind of like organizational setup. So for example, we have organizations who's operating in very many countries and actually every country has its own like finance organization, legal organization, HR organization. So it's multiplied. So it doesn't make sense maybe to join forces and to have like a center of excellence or having, global process owners? So this is also coming along here within your investing in the technology infrastructure. Also think of what you wanna see in terms of what your organization. Then the next one is pinpointing the data stewards. We have seen it often that after the implementation, the organization might get a little bit lost because now we have all of those great new processes in place and we have also identified, process owners, within, for example, Workday here. But what we have seen is that, of course, Workday as your ERP system is also connected to other systems. So really making sure that you also have, like, your data stewards. And also in terms of AI, making sure that, you really understand the use cases. So, first step is, of course, to have your road map, what you wanna achieve, and then the second one is also we have your AI use cases. And then you are deploying those AI use cases, But then also what's happening with them. And with the data stewards, it's a great, kind of, like, role that is really kind of, like, understanding the I I AI use case, but it's also updating it because they are kind of like the world is changing every day. So does it mean that we wanna use the AI use case that we have to call it or do we also wanna invest and update them because the world is also changing around us here? And then getting also better understanding and connecting with other organizations, because what we have also seen that systems are connected, of course, and each of the different systems and solutions bringing another value, to the organization. So stepping up here and also connecting with the different organizations. And the next one is deploying agent. Identify identity profit pools. And that's really kind of like a game changer what we have seen. And that's coming back. Okay. Reviewing where you have, in terms of weaknesses, spending a lot of time. So actually, we have clients which are reconciling all the different sub ledgers and also the the banking connectors. Also with your, for example, your operational system reconciling all the data, to kind of like make it fit that all the numbers are putting in together, but then really using AI in order to make this work for you and also, identifying based on the figures which has been generated profit pools that align with your strategic focus. So, actually, bringing in your view on your cost structure, but also bringing in the re of your revenue. And then it also comes or connects again to the trends that the office of the CFO was not no longer there in order to provide all the finger figures for the reports for the c level, it really becomes those strategic business partners within the organization. Because we have the data available, you can compare. You're not spending hundreds of hours in kind of like making the figures fit. Actually, this is supported by AI and it's also kind of like helping you to understand what you can actually do and where you can improve as an organization. Having said this, and before we go into this slide, a quick wrap up what we have seen, in the trends. A lot of trends were coming from, how the world is changing. So that means that we as an organization also need to adapt there, which is starting from planning and reporting, which we have seen that the talents are changing and also the finance leaders are changing, becoming more strategic partners. We also see that AI really can help us, especially in the areas where we might have spent a lot of time. So Agentic AI can help us stay a lot. And this will also help us in overcoming the ROI, which we have seen a lot of organizations are still struggling with. And then handing over. Paul, you might be on mute. Are you? Oh, sorry. Can you hear me now? Yeah. We can. Alright, okay apologies. Okay, so what I'm going to quickly try and cover here is, the work that we're doing in Workday around AI and agentic capabilities, and give you an example or a flavour of what we're doing in each of these areas. So if we're gonna just move on to the next slide. So starting with with FP and A, I guess with the, you know, the the decline that we're living in at the moment, the macroeconomic uncertainty, geopolitical issues, You know, companies can no longer really afford to be planning on a quarterly, even monthly cycle at this stage. You nearly need to be able to have that agility to plan on a daily basis. So obviously with, you know, these are these uncertainties can create both opportunities but they can also create risks. So being able to kind of plan various scenarios very quickly to be able to seize those opportunities or mitigate those risks is you know as important as it's ever been. So if we just move on to the next slide, so one of the agents that we, are working on is our planning agent. So this is in our, you know, adaptive planning, area. And there are many roles involved in the in the whole planning process and maybe we look at what some of those roles are. But essentially, the Planning Agent that we're creating, will have skills depending on the role that you play in the in the planning process, it will have skills to support each of those roles or those personas. The agents, you know, Maureen talked about some of the skills and capabilities of agents. So one of the things that's really, I suppose, important, is is having that contextual awareness around the work that's being done. So the agent will know, you know, what what role you're playing in the planning process and have the skills and the awareness to, know what area of the application you're in and what type of data you might be working with. And the last part there is around how you integrate or how you interact with the with the planning agent, and that's basically through natural, plain English. So part of the planning process can be quite complex. You can have very complex formulas that you're that are involved, as part of the the the forecast and the plans that you create. But you just interact with the agent in plain English. So if you have a very complex pricing formula, for example, you can just type what you what that pricing formula needs to do in plain English, and the planning agent will create that formula for you. I'm gonna look next at some of quickly some of the the roles that that we, that are played in the planning process. But we're just gonna focus on on some of the skills to support the analyst and the planner here. And, you know, down the line, then we're gonna have focus on more skills for the for the modeler role and the admin role in the planning process. So first skill I think we're gonna look at is around the scenario planning. I don't have a demo here, but, essentially, what the scenario planner will do is allow you to come up with a a scenario, that you maybe wanna play out and see how how we could plan around that. In this case, you can see in the middle there, we're having a conversation with with Workday in terms of being able to for this this is an airline, customer, and they wanna look at opportunities for increasing the revenue by 10%. So what you know, rather than just, you know, obviously, do all the calculations around what the revenue increased revenue numbers would be, they will actually come up with a complete strategy as to potentially how you could increase your revenue by 10%. And that encompasses things like looking at the pricing adjustments there, looking at capacity planning. So whether that's looking at, you know, routes that maybe not maybe less profitable that you might want to reconsider redirecting different routes or the capacity and, you know, seats, occupancy on on flights. So it's looking at all those types of scenarios. And even looking at things like maybe customer and loyalty programs and how you can maybe tweak those or improve those also to to increase revenue. So it it just it gives you that complete options around creating, creating that scenario. And then just the next skill I wanna quickly talk about is around variance analysis. So being able to kind of understand you you can essentially create the the parameters in which you want to really be notified about any sort of, key variances. So you can set thresholds or limits and have the agents, you know, constantly monitor those. And then it can surface it'll surface those variances. It will give you the root cause analysis around those. But if you need to summarize that and and bubble that up the organization, it can also put into plain English a an executive summary around around what those variances are, and also some recommendations to alleviate the those in in the future. So that's a little bit on the planning side. We're also doing a lot of work on our insights, and essentially being able to have conversations with the data. So, in this particular example here, I mean, we have a demo around this one as well. You can have, you know, rather than kind of creating reports, this is really you can maybe look at this example here, flux analysis. So you might want to interrogate surface queries insights of variances and then also get that deep root deep cause or deep root analysis around it. I think we'll just maybe run the demo here. It's for a resort organization. They they have multiple resorts. They have, obviously, rooms, but they also do banquets and and functions. So we wanna do a route, a flux analysis of comparing month over month. I'm feeding up my rules, so the thresholds and the types of variances I want to be notified about. And it's now analyzing the data from my, across all my, resorts. So the only one that's flagged here is around our food and beverage. So this is around events that we host, banquets, for example. We can see here that the variance is 58,000 for the month. The threshold is 50,000, but it's now analyzing what might be driving that. And in this case, we can see at the bottom there is a blue line that's saying that we've increased the revenue for our banquets and functions, but we've also increased the cost of running those. So those banquets are kind of fixed price, but the very the food to run those, had a spike, so it had an increase. So that's eaten into margin for our events. So it's looking at that across all the organ all the the resorts. You can see that it's this is an issue across all the resorts, not just say one or two resorts. And it's also now looking at producing an executive summary. So if you want to bubble this up for senior leadership, it will give you that a root cause, it'll give you the executive summary and it'll also give some suggestions or recommendations how we could look to alleviate or reduce those variances going forward. And then the last thing I just wanna quickly talk about in this particular area is, you know, our work that we're doing to help surface insights. As Sina kind of mentioned, finance leaders are being really challenged to become, more strategic partners to the business. And how they can do that is really by being able to surface quickly surface those insights, that can help make help the business make faster and, better decisions going forward. So in this particular example with our reports insights agent, you can have a conversation with a report. So you can generate a report, but rather than, having to extract that report maybe into Excel to pivot or filter to get various information, You can just ask questions, to ask Workday to the agent there. Maybe you've got a list here of supplier contracts, and I just wanna know what's the total value of all my, you know, current or open, approved contracts, Or what is the total of all the contracts I have with a particular supplier, Dell, for example? So you don't need to take the data out, do those pivots or filtering. You can just keep the data within Workday and quickly surface the insights that you need. Okay. Moving on again. So the next area I wanna kind of focus on quickly is around profitability. And so one of the things that we are building out shortly is our CPM, our cost and profitability management, tool. And this really allows you to pull data sources from your financial GLs, but also data that might be coming operational data that might be flowing through our accounting center. And that can also include the drivers for some of the, profitability metrics that you want to produce. So the idea here is that you can get more granular insights into profitability. That may be profitability across your, you know, line of business, different, SKUs or in you know, or different, products or customers. So you're able to slice and dice, that information as Sina talked about earlier as well. You know, getting those insights into the profitability, across the business is really important, but being able to get that granular, view. And I think we have a little demo here as well in terms of how our CPM tool, will work. So this is kind of the profitability analysis hub. You've got some of your key KPIs at the top there, like, around income and operating profit. This is a bank and then we can drill into their different products. So they've got consumer banking, commercial banking, we can see the revenue and the direct costs. But we can also see then those costs have been allocated, those overheads across the business, and see then the post allocated profits for each, product line. But then you might want to kind of come up with a different driver for allocation your IT costs. So where where we can help with that is, you know, using AI and the Agennta capabilities within within this, product. You might wanna go and put a different, formula for allocation your IT costs. So you can just kinda, ask Sana here, look. I'm looking to allocate my IT cost against my virtual CPU hours. So now you can feed it an example of what that might look like. So you can pull in CPU hours for each of those, departments and then also kind of feed it in what your company policy is around cost and and profitability. So you upload those files. So that's given the context and an example of how you want those costs allocated. And then the agent will go off and will kind of take that sample data. It will now work out based on the CPU hours or the virtual CPU hours that would be used by consumed by each department, how those IT costs could be allocated across, the various product lines. And then if you're happy with that, we'll actually create that rule, create the allocation definition so that in future months when you're allocating your IT costs, it's gonna base it on those CPU errors that that you feed it. It'll allow you to get that granular, insights across the business and help determine if there's maybe certain products, product lines that just aren't as profitable and maybe not worth investing as much in the future. And then keeping on the theme of, profitability, we also have our contract, revenue contract agent. So we acquired a company about eighteen months ago called Evisort, which uses extractive AI capabilities to really look at, you know, very lengthy, worthy contracts or documents. We've been able to act extract the key, data, that's buried deep in a lot of these contracts. And where we've utilized that those AI capabilities is around, managing revenue contracts. So if you've got thousands of revenue contracts that you're you're managing, there's a lot of critical insights that are buried deep in those contracts. But we also wanna take that a step further. So extracting those key, terms and conditions out of those contracts and actually creating those contracts within the Workday, system so that they don't have to be done manually. And then down the line, we want to actually create more of those downstream transactions. So once you've created your revenue contract, we could help with creating the revenue, the billing schedule or the revenue recognition schedule, and automating the creation of those, contracts. But for customers that have been using this already, we've seen, you know, 60%, reduction in errors, because we're automating the creation of these contracts rather than keying the main humanity from from those, contracts. And that's an increase in 35% of the time, increased time to to billing readiness. And and here's just a quick example, of a sales order on the left where we've extracted the key terms, but we've done more than that. We've the the agent's been able to reference back to a master service agreement with this particular customer that references various volume discounts. So now they wanna order place an order for a thousand units at $250 per unit. But the according to master service agreement, once they've, purchased over or ordered over 1,000 units, the unit price should drop to 230. And they've previously already ordered a thousand. So for this particular order, that we we should really be readjusting the unit price to reflect 200 units at $2.50 and then the remaining 800 at $2.30. So it averages that out. For this particular order, It's recommending that we use the unit price of $2.35, for those thousand, units. So it's it's been able to reference back that master service agreement, propose, an alternative unit price, and then actually create that contract in Workday for you. And then moving on to working capital optimization and and really kind of a couple two examples here. So one is really focused around streamlining, processes. So we're making them more efficient and helping to increase the working capital that way by efficiency. But, also, we look at our supplier contract agent, similar to the revenue contract agent, making sure that we're maximizing the terms that are buried deep in supplier contracts. So on that on that note here, so with our supplier, contract agent, we know that, like, we know from our research that, nine percent of, approximately of annual revenue is lost, due to per contract management, so not being able to keep on top of all the contracts, supplier contracts that you have, that we're not realizing up to 40% of the negotiated savings that you've negotiated that are buried in those contracts because we're not keeping on on top of those and not servicing those those insights. And then in managing manually managing contracts, the contract life cycle, it's on average wasting three weeks. So just an example here with the supplier contract agent. There's a a little dashboard on the top there where you could you can see there's an example here that says you're you're you're eligible for a rebate. So you've you've built into a contract that once you've spent over, a thousand, euro, in in the quarter, you're entitled to a rebate. Now unless you're manually reviewing those contracts all the time, you're probably not gonna be aware that you've, you know, that you've hit or you're entitled to that rebate. So we're surfacing those types of insights because we've consumed all the supplier contracts, and we're constantly agents of monitoring that, surfing the surface surfacing those, rebates or insights, and then having dashboards that can tell you things like, you know, when maybe certain contract terms are coming to an end so that you can get ahead of that contract negotiations. And I think the last thing I just wanna quickly talk around about is, you know, around the, efficiency, aspect. So one of the things that we will that we have or working on is our cash application agent. So currently, cash applications is, you know, is is automated to the extent that you can take a bank file in. If that bank if that payment on the bank statement is just paying maybe one invoice under the reference field in that on that in on that bank statement has that invoice number. Yes. We can allocate that payment against the the customer invoice. But often, you know, there's multiple invoices being paid and the reference won't hold all those invoice numbers. So now with our Cash Applications agent, we can combine consuming the bank statement information but also the remittance advice information that will have come in via email, into the organization. Using our extractive AI capabilities, we can extract all the key information from each of those remittance advices, populate them into Workday. So now we're combining the remittance advice information plus the bank statement information to make sure that we're able to apply all, you know, the cash that's been received against all of the, correct invoices that are being paid, on on this particular payment. So I think I went through that pretty quickly. I think we only got five minutes left. Apologies for overrunning, Marion, but, hopefully, that's giving everybody a flavor of the the types of, agentic work that we're doing at Workday. No problem. Thank you, Paul. And we just before I hand back to Sina, just to kinda bring us home in terms of the report, I wanted to highlight just the topic of AI governance because I know that this is, you know, a pretty top of mind for a lot of people who when when they look at some of those agents that we're building. So if we look at the next slide, these are the three foundational capabilities, and I touched on these earlier on, but, essentially, that every AI agent in Workday is built on. So context. So ensuring that every agent interaction is constrained by Workday's real time security model and your org structure. So they'll only ever see and only ever act on what that specific user has access to. And the second is rank configured compliance. So they inherit the business processes, approvals, policies, and regulatory rules that you've already, built into Workday. And then the third is scoped execution. So they execute with the exact scope of the human that they represent. So there's no super kinds, no excess privileges. Every single action is auditable and traceable to who did what on what on whose behalf. And then if we look at the next slide, we, you know, we hold every agent the same security as a standard human worker. So every agent must have a unique identity provisioned in the platform, no shared accounts, no anonymous bots. Every agent must authenticate to establish their identity, and then every agent must be authorized to access specific resources within the product. And the last one then is all activities logged and recorded, you know, creating that audit trail for the agent. I think that's really important, and it kind of ties into that area that Sina was talking about a few minutes ago around, security concerns. We know that this is top of mind, and this is why we are building and reinforcing that security model within our agent framework as well. And just to highlight, you know, we've built AI fact sheets. They're on the community, so every capability and every agent that we build essentially has a fact sheet. And that, you know, is designed to give you a clear detailed understanding of the AI models and the data that they're using. So kind of think of these as a nutrition label for a food product. Our AI fact sheets provide a similar level of transparency for all of our AI solutions. And I just wanna hand back to Sina for a second. I I know we're kind of coming up on time, Sina, but maybe, Ian, if we can jump to the next slide. You know, I think talent is a a real important topic here, and, I know Deloitte's report has, some key trends that you wanted to highlight, Sina. Yeah. Sure. Thanks, Marion. Yeah. We based on the report, we have seen that really sixty four percent of the respondents shows at least one technical skill that they wanna develop through 2026. So what you can see here, all the top skills developments, areas, which we typically see in the finance organization, Whereas you can see in the green highlighted are the technical skills. Like, of course, there's AI included and automation skills or data analysis skills. Whereas we see that, of course, that, all the basics like the human skills are still important and needs to be developed, but also the business skills, are still something that we all wanna work on. But I think it really points out that specifically technical skills are requested and that a lot of, employees wanna work on. So it really comes back to the trends we have identified that the finance organizations, it's not only there to provide the figures, it's really becoming a business partner. And with this also kind of, like, giving a twist in the CVs here. Super. Thanks, Zena. So that is, the basically, the trends, from the Deloitte report, and really delighted to have Zena to share those with us. This is just a couple of, you know, kind of key takeaways. You know, we identified that value gap, and it's really a design problem, not an AI problem. You know, you can't bolt agents onto broken processes, and ROI will really come from redesigning the work and redesigning the processes and especially taking into consideration your people, in that process as well. Agentic AI is only as trustworthy as the platform beneath it. So we talked a little bit about context guardrails, audit, and security. They'll have to be built in and not stitched on, at the end. And third, you know, start where the cognitive load is highest. So Deloitte's research points to three domains where agentic AI pays back fastest, FP and A, sales and profitability, and working capital. And, you know, for all of our Workday customers who are on the call, Paul highlighted, a number of agents that we're building in this space, you know, kind of aim to pick one, improve the value, and scale. And I we have, like, a minute left, Sina. I don't know if you want to do a quick, you know, point to the audience. Anyone who's, you know, genuinely looking to get started, you know, what would you kind of what piece of advice would you give them maybe, to start over the next ninety days and where to begin? Yeah. So first of all, I think the pre step would be checking in in terms of if your basics are correct, your, systems architecture, and also where your data is coming from. And if you have done this and you really kind of, like, wanna keep dive into the more AI space, I really would recommend reviewing what are your, we have more weak areas or where you're spending a lot of time where you kind of, like, really have to sit there, do manual work, and then reflect new case. This is something that actually can be supported by AI, which in the end frees up my time and it's bringing up more value so that I can focus on really helping the organization maybe to grow or maybe shifting into a different direction, bringing in new ideas here. So that's really what I wanna focus on. Super. Thanks, Ina. And thank you all for sticking with us for the past hour. I hope you found this helpful. As I mentioned, a copy of the report can be found in the docs tab if you wanna deep dive further. We have built a, a short a short report with Deloitte that we're gonna be publishing soon, so we will share that after the webinar, and we'll also share the recording with you as well. So with that, thank you, and enjoy the rest of your morning or afternoon, depending on where you are.