Ep 05: The Agentic Procurement Diaries: Fifty Agents and a Stale SharePoint: How DXC Built Its Own Procurement AI

0:00Hi there, my name is Philip Eidson and welcome to the Agentic Procurement Diaries. It's a show where we explore what procurement could become in an age of Agentic AI. The Agentic Procurement Diaries is presented in partnership with Zycus, a leader in Agentic AI for autonomous procurement. If you are a CPO navigating this shift, this show is for you. The truth is that nobody has all the answers yet, not even the most forward-thinking leaders. We're all learning, experimenting, and figuring out what this technology could mean for our teams and our function. In each episode, you'll hear from CPOs about what they're experimenting with, what they're learning, the challenges they're encountering, and how they're preparing their organizations for what comes next.

0:42These are conversations about the journey told by the people taking it.

0:50Today on the show, I'm delighted to welcome Anders Lange. Anders is the Chief Procurement Officer at DXC Technology. It's a firm that helps large organizations manage, modernize, and secure their IT systems. First of all, Anders, welcome to the show.

1:04Anders R. Lange: Thank you so much, Philip. This is a pleasure to be here. Talking about procurement in a setting like this has been my lifelong dream, or at least since I started my career in procurement. Well, let me ask you, first of all, welcome. I'm so happy to hear that. How did you get yourself into procurement? Well, that's a great question and it's a question that once in a while I ask my team about. And nobody said I want to be a procurement person when they grew up. And maybe not even in college was it their first plan. So I end up in procurement, kind of like two paths. One, I realized afterwards was a perfect segue into procurement, which was my education.

1:47Anders R. Lange: So I had a degree or have a degree in commercial law and business administration. So a business school education, but five years of starting law. And there's a lot of contract law and negotiations in that. I didn't realize what procurement was until I had an internship in the States. And my boss at the time said, I have this friend who works at Carlsberg. Everybody in Denmark knows carl spook a lot of people in the world i say cool i'll do whatever you want for them and it just happened to be procurement and and when when i when i looked up that word and procurement procurement is not a word you use in denmark where i'm from um i said wow this is a perfect match and and since since you know studying and and you know getting out of education i've been in procurement yeah no you said that not everybody wakes up thinking they're going to be in procurement i I usually ask that question in my podcast and you write for like 99.9 % of the time.

2:49And could you talk a little bit about kind of the, your role at DXC, what's your kind of, what procurement, kind of the purview of procurement and how it fits into the broader organization?

3:02Anders R. Lange: Yeah, absolutely. And in fact, we call also supply chain management, although we don't move a lot of goods and logistics is not our main service line, if you will. But really it encompasses everything that touches third parties or suppliers, the typical strategic source, and category management functions, the procurement operations, and all the enablement of suppliers, but also accounts payable. We have functions that track our discounts. We have functions that support mergers and acquisitions and the restatures and whatnot, as well as corporate services. So travel and expense, it resides under my umbrella.

3:49Anders R. Lange: So does the logistics piece that we do do in DXC. And as I recently also real estate and facilities is under my umbrella. So not the typical chief procurement officer as a role, but all the ancillary functions that a lot of us deal with many times. Surprisingly, to some, maybe risk doesn't sit with us as it used to, but it sits within legal. So it's a typical procurement plus AP plus corporate services, I call it. It's quite a remit. Like a lot of those functions that sometimes fit within procurement, all of them are fitting within your remit at DXC. And can you talk a little bit about kind of the, I was going to ask about the culture of procurement, but I'm thinking about how procurement is perceived across the rest of the organization, you know, and kind of what you play in, what you might not play in, just for a little bit of context, kind of before we start to talk a little bit more about AI.

4:50Yeah.

4:51Anders R. Lange: And I should mention that we sit under the finance and birth, so under the CFO. And a lot of what we're tasked to do is the traditional cost control or cost savings. things. But when it comes from a culture standpoint, I've been with DXC for 18 months and one of the things I have instituted or continued, if you will, is really around transparency. So understanding what we do as an organization, how do we support the business? What do we do? Who does what? Typical pipeline that most people will deal with is a key thing, but also accountability as that. So who's then accountable for which projects, which form of the pipeline so that we can drive that, let's call it business alignment.

5:42Anders R. Lange: And in all of that, make sure that we are completely connected with the business that we support. And when I say completely connected, obviously, you have to work on the relationship. You have to work on getting that famous seat at the table. But by understanding what's on our plate, we can better address those conversations about getting involved, making sure that we influence spend and all the things we expected to do. And that kind of goes without saying, I think in many organizations, that's what you want to do. But I think it's so critical you to emphasize relationships and building those alignments with the business.

6:23Anders R. Lange: And we'll probably get to that further on in the conversation when it comes to AI and and whatnot. I think relationships become even more important because you can have computers or technology or AI take over the world if you let it, but we still have people in the loop. So that relationship-driven approach and being able to build that trust with the business is absolutely paramount in my opinion. And I think that's where I've seen a shift certainly in the time I've been here is that the business wants to work with somebody who cares, somebody who takes accountability, somebody who uh um you know ask how can i help um and and it's actually fairly simple it's just having that you know interest and saying we're here to help that uh that certainly promotes uh you know the business wanting to come back and work with you yeah i think that's as as much of a mindset uh isn't it on on the entire team and thinking of yourself as uh i always think about as you're kind of a business within a business.

7:26And so you have to provide a value proposition that's compelling enough for the stakeholders around the business to want to come back and work with you. And actually, when they think they've got a problem or a challenge or an opportunity, they think, I think procurement can help me with this. That's kind of the state you want to get to, isn't it?

7:44Anders R. Lange: Oh, no, absolutely. And again, if you think about that, what a great position to be in, right? you know, because procurement as a function of optimizing, you know, third party relationship and optimizing meaning, you know, not only cost savings, but, you know, it can be the quality of what you're getting, it can be the time you get it. And being able to add something to an engagement that either your stakeholders are too busy to do, or they have to maintain relationships or otherwise not able to do that, you don't find any stakeholders don't find that valuable. If you do it well, you also have to do it oftentimes at the agenda and time that the business needs you to.

8:28Anders R. Lange: So procurement is a tough, but it's also a very rewarding one when you get it right. Yeah. So could you talk a little bit about where you are on your journey with Agentec AI? Are you just exploring? Are you piloting? you further along, where would you characterize? So we are certainly beyond exploring and pilots. I did a review with my team the other day as to where are we, and we have more than 50 active agents. Now, what does that mean? An agent can be something that is simply helping process data from one process step to the next one. So just some examples for context, we have one that helps us with tailspin negotiation.

9:18Anders R. Lange: So did the things we didn't have time to touch before. It just took too long to look through a code, derive a response, and then send it back and maybe go to a balance of no, I want this code to be low or I want the payment terms to be better of what it is. We have a contract review, a contract analytics agent that tells us, okay, this contract compared to our standard terms presents these risks, which I think are all like table sticks when it comes to a lot of these agents we have in our space at this point. Or certainly there's a lot of use cases you can see from different providers that provide these things.

9:56Anders R. Lange: So we have those, I think, fairly common use cases addressed as agents and continue to develop those. The one I would say that I'm the most proud of where we've invested the most time is one we call the supply chain advisor and really just as our extension of our team, which is starting with the fact that we got a lot of questions. I got a lot of questions about where is this PO? Who is this with? What is the status of this thing? And so forth. And we get it at all levels. and who do I engage for this or what is our policy on that? And by and large, we had pretty good documentation, but they sat on SharePoints.

10:39Anders R. Lange: They sat on different webpages. They sat in different folders that were more or less easily accessible to people. They certainly could if they had the time. They knew where to work. They knew where to look and found it interesting to go through layers and layers of supply chain and procurement webpages So we started with, we need to be able to give end users an easy experience of just asking those simple questions. So we really tied this agent back to all of our supporting process and policy information. And then we tested out, we've asked at this point probably more than 500 questions to see how well does it actually answer these questions that we commonly receive.

11:22Anders R. Lange: And it wasn't great. And in part because we learned that a lot of the information that we had on our web page was stale. or some tweaks had been done to a process or it pointed to the wrong folder that might have been moved from one place to another. So we spent quite a bit of time to make sure that the information was up to date. And I think that's as we progress in the conversation, the Data Foundation and the Process Foundation I think is absolutely paramount to making most agents useful and achieve the outcomes you want. But once we've done that, we start to see now in the tune of 500 per month, we're probably saving on not answering these questions, getting people to the right point.

12:02Anders R. Lange: To where we're now using this agent also as our intake and I would call orchestration, where rather than having people go into our existing solutions, which people they are familiar with, so people who do it for a long time, they'll do it, not willingly, not happily, but they do it because they know that's how the process works. But for new people in particular, you don't have to overcome that barrier of adoption as much because most, if not everyone in our organization are used to using some level of AI to ask questions. So now it's just using that same type of approach to get information. And now also, what happens after I raise this purchase requests or I approve it now with an extra provers I have to go chase if that's what you want to do so so so it becomes an intake where we can we can ask people to go you know put put your questions or your request for support in here we tie that to to our existing you know intake process and and then and then the entire process from from really you know I have I have need to make a payment.

13:20Anders R. Lange: So that orchestration, I think, is super exciting because I know how much it costs to put orchestration in as its own solution. And the reason why those tools are very popular in our industry right now is because they help adoption, they help the process go along, and they have a lot of value. But in DXC, we're kind of developing that internally through the needs that we have. And I can talk a little bit more about why we decided to do it internally. Did you find that, like, how did you bring all the data together? You talked about data being critical. And I think you don't necessarily realize that until you start building and you start to see some of the outputs of where the needs are.

14:08but you talked about it being on all these disparate locations and people have to search for information. Did you then look at that and say, I need to create a single database or a knowledge base or a data lake or whatever the right kind of thinking in terms of terminology is? And that became the foundation of what you were then able to build the agents on top of.

14:27Anders R. Lange: Yeah, and that's very much the thinking but also very much work in progress. So if we think about where we tie our certain features features of the agent too. So for example, being able to look up your purchase request, make sure that that information is in the data lake. Now, if you want to do it with contracts, and it's another instance, another system, then we have to, so our plan is to effectively, you know, create a data lake for all of these data points that we want the agent to look up, if you will. And that's probably our, you know, the agent is kind of the front end or, or the exciting things to talk about and show people, but the hard work and the foundation is built upon making sure that we have the data and the appropriate data lake.

15:15Anders R. Lange: And as I said, that's very much work in progress. And every release that we have in feature of our agency is then tied to what can we do with our data and what is the next step of that data lake journey, if you will. Yeah. Yeah. When you were starting to think about how do we leverage agentic AI, you know, was there a moment where something that happened or you saw an output or just something where you thought, you know what, I can see how transformative this can be. It's like the aha moment. Did you have one of those moments? Yes. So I think there are kind of two things that happened. And, you know, I came from an organization that didn't have necessarily the same need or were not a first mover when it came to AI for different reasons.

16:03Anders R. Lange: And coming into DXC that, you know, have been on a long transformation journey. IT service is notoriously a transformation. That's kind of part of what we are selling. And our CEO is like, we need to leverage AI. So early on, it was, you know, what tools do we have available between, you know, Microsoft CoPilot that most people have access to in some shape or form. and that push to see ourselves drive more value for clients through AI. But then as we then start to see, okay, where can we pick that up? And is there an opportunity to find opportunities to leverage AI within supply chain and procurement, not only for ourselves, but also for client kind of thing?

16:49Anders R. Lange: That was that moment that hit home. And then starting to use, we use Amazon Quick, which is not widely recognized maybe as one of the typical AI solutions, but Amazon is obviously a well-known company. And seeing some of the outputs of what we could get through that desktop version, you tie your different folders into this agent. so the knowledge base becomes, you know, for contracts, for example, and then asking, you know, what are some of the recommendations, you know, there is for this upcoming contract negotiation, and to see the level of detail that's been out and being a procurement professional, where you know how long it takes to do this read through a contract, let alone identifying the opportunities and seeing how it takes your data coupled with whatever other sources the LLM have access to and come up with some great tangible suggestions that we could even validate with third-party market intelligence.

17:58Anders R. Lange: That was amazing to see. And not that we couldn't do it, but it was the time it took or would have taken us to do that. And the things we would have missed either because we are pressured on time, you know, on internal deadlines, or because we're humans and we tend to focus on the three, four, five most important things. There is a laundry list of things that we could address, which came very handy in specific negotiations because, you know, I don't get these five. There's other five. Yeah, he's a helps with the planning. And I think that ability to not just, you know, finding an answer to a question, quicker, the ability to do the analytics, bringing so many different documents together.

18:43Anders R. Lange: You said when we chatted before, you've been a long time in procurement, so have I. Some of these contracts, you can have 20, 30, 40, if not more versions of a contract. Being able to find sense and what to focus on within that landscape, I requires you to have been part of all those 20, 30, 40. You probably wrote them. when you came from your contract background. There you go. And nobody has that luxury of remembering everything. Again, you're only a human being. So the way that AI can supercharge that analytics and the memory is incredible. Yeah, it sounds like at DXT, you've got a kind of a top-down culture of adoption of AI.

19:31have you had pushback or you know concerns it could be within stakeholders it could be within your team it could be the executive level I would love to understand kind of any roadblocks or where you've believed like this is something we really need to do and you may have come across

19:48Anders R. Lange: resistance yeah yeah I mean there has been surely one of the reasons why we do it internally is also cost because it costs a lot to go and acquire technology. In part, it's also, we have this customer zero concept and see, can we develop something our customer actually will want to use? But that has also come with constraints. We have to use our own team. And we have a great team, digital procurement or systems and technology team that can support that and not everybody does. But that's a constraint. And where the resistance comes is, do you need to spend time on developing these things yourself?

20:30Anders R. Lange: What about all the typical things between data privacy and legal and IT? Now, developing it yourself means that having, whether it's Copilot or in our case, Emerson Quick, the data that is in RingFence is in our firewall. And that made a huge difference where we did do pilots outside of DXC, we ran into to the typical challenges of data privacy, of testing these applications, understanding where the data went, checking our contract as to, can we actually put this contract into this agent because where are they going to store this information or how are they going to use it? So those typical things were issues early on, which is one of the reasons why we went down the path of using our internal tools to then create these agents.

21:21Anders R. Lange: and that has certainly propelled us in terms of the development. The other thing is that many people ask is, how will AI impact my job? I'm used to doing this function of moving things from one process up to another or like I used the example of contracting before. Now you can do that a lot quicker. People are starting to ask, you'll be, you know, will I still be here when AI is done with its thing? And one of the things that we keep, you know, pushing as a company is this human plus or human in the loop, in part because we've seen in cases where, you know, you trust AI output fully. That's, you know, I wouldn't write it that way.

22:09Anders R. Lange: I wouldn't write the email that way. Or, oh no, it's suggesting 90 day payment terms when our contracts are 120 days. So I forgot to give it that context. So just having that human loop is critical to us. And I think that is what we stress with people as to, we have to change how we work. We have to adopt AI as another tool in our tool, but a very effective one. But do that with still having judgment, understanding the output of AI has to be put in the context of the question, the problem you have to solve. So, you know, that has been a big one. You know, how will AI impact my job? And again, we're trying to stress the fact that AI will impact your job.

22:53Anders R. Lange: But if you see there's an opportunity more and more as a threat, it will have long-lasting positive impact on your careers, whether at DXC or somewhere else, frankly. Yeah. No, I think that whenever there's any big change, we talked off mic about outsourcing a little bit. I've been involved in outsourcing during my career as a practitioner. And, you know, my view was always, I don't mind outsourcing my job because that's going to give me new opportunities or there's going to be, you know, 5 ,000 other companies who want me to go and do the same thing for them. And so I always felt kind of some level of job security amongst doing it.

23:29And I think now with AI, it's, I mean, the same is true. and you either you embrace it and you take control by doing that or you kind of become hesitant and that's when it's out of your hands and that's what I tell folks, like you want to be in control of your own destiny, then you need to embrace technologies like this

23:50Anders R. Lange: No doubt and I don't think anybody can apply for a job certainly not in procurement without there's some mentioning of artificial intelligence in the job ad and without you being able to explain how can you adopt AI, like any other technology that came before it. So I think you're absolutely right. When it comes to the role of the human, you talked about human in the loop, do you see that there are areas where you would trust an agent end-to-end and not have a human involved? Or that even if it's just a light touch at the beginning or the end of a process that really, at least at the moment, that a human still needs to be involved at some point in a process?

24:36Anders R. Lange: Yeah, there certainly are. And there's some things we're exploring between, it can be a supply categorization or spend categorization. We know based on our data what are some of the categories of spend that is easy to categorize. There's a high level of accuracy with the data. Whereas others might not be. or we have a certain mechanism of applying payments to certain of our invoices that will allow us to get some discounts. So how we categorize those is highly rule-based and kind of contained within our function. So nobody has to see how the source is made, so to speak. And we can control through either checking up on percentage of what it's validating or what it's processing without breaking something or sending something to someone that shouldn't receive it.

25:43Anders R. Lange: So we're starting to do that already, kind of contained within our own process or, in other words, if there is an error or something happens that shouldn't have happened, we will be the kind of recipients of that issue. None of our stakeholders, whether internally or externally, would see that. One of the things that I see procurement leaders trying to kind of figure out at the moment is this balance between opening up the AI tools to their teams to develop their own use cases, to develop their own kind of agents, if you will, that are within, that help them do their job more effectively, whether that's faster, whether that's, you know, in a way they couldn't do it.

26:32do it before versus needing these controls at a central level to make sure that you don't suddenly have 500 agents of which 400 is kind of doing the same thing. You know, all kind of out there with a lack of control. How do you think about kind of balancing? You want to give your team exposure because that's where they're going to develop some of these, you know, the confidence and the trust and the skill sets that you need while maintaining some control over kind of this is what is approved and this is what's not

27:01Anders R. Lange: approved. No, we were all in on that concept. We even had competitions across part of the company of who can leverage AI and develop the most agents with the biggest impact in part to recognize adoption as it happens to use and solve a use case, but also to give people something. The new tool to help fix some of the problems that happen. So again, with AI, you don't have to necessarily rely on IT others to set aside time for you. But also recognizing what you said about you have several agents doing the same thing. And we've gotten to that point where we have had to institute governance and we have now central database or dashboard with all our agents, because people were doing not necessarily always the exact same thing, but some level of supply intelligence.

27:54Anders R. Lange: What does this contract say? What does your supply data say? So that now, Now, not that people can't create a new agent, but you go to this library first, you know, here's the owner of this agent, and talk to this person along with our digital procurement team to make sure that we don't, you know, reinvent the wheel. Because not that there's a lot of problems, challenges, and opportunities in procurement, but there's only so many that haven't been thought through already. And I think we're reaching that saturation within our group around, you know, what new problems are there. So we have put that together and also to avoid people, you know, wasting their time.

28:33Anders R. Lange: But also I think more importantly, you can create these agents, but the throughput is low and or so how many things do you put into the agent or the output is low. You might put a lot of in there, but the time saved is very little because you have to review it or whatnot. Or you're not able to scale it. So maybe only a few people can use it because, you know, it only supports one language or whatever it is. It's based for a very specific niche use case that fits an individual's working style as opposed to something that could be scaled more broadly across the team. Absolutely. So to make sure that it has broad uses.

29:08Anders R. Lange: And in some cases, can it be used by someone else? So we do share across our organization with our finance functions, with our support functions. What are these agents able to do? because if we talk about translation, a lot of have problems with translating an invoice from one language into another, for example. Every function has some issues like that. So I think that governance is key. But also I think speaks to the majority of our AI journey or roadmap now is that we have enough agents where now that becomes a challenge we have to do. But I think it's a positive one. I rather want that challenge than not figuring out what agents to implement.

29:47I have a couple of questions just before we leave. First of all, just on the journey that you've been on so far, are there any key pieces of advice that you'd like to leave listeners with?

30:01Anders R. Lange: I think making sure that you involve your team in this journey as to what problems do they face. Give them the ability to innovate, come up with ideas. again, within the Gondra, as we talked before, I think it's absolutely paramount because I think there's a lot of fear and uncertainty with what AI, but as I said before, I think it can be the greatest thing that has happened to our function, but also, again, that uncertainty that plays into negatively. So making sure that there's that ownership and participation, I think, is key. You're still maintaining close relationship with IT, data privacy, and legal when you put this in place because we not only do it for ourselves and our youth, but we obviously support the business in getting AI introduced to the company and we don't build everything ourselves.

30:51Anders R. Lange: So I think that early on, which I think everyone will recognize when you buy software, buy anything, technology is bringing the right stakeholders so you don't move too far down the line and then you get disrupted when you're too far down the line. And then I think just be nimble and adaptable. dealing with software or AI and the suppliers that provide that who want five-year contracts. If you put that in context, just go back and see how many versions of, take any AI, whether it's Amazon or OpenAI or Anthropoc, how many versions of those AI tools have come out? Yeah, it changes so often of like which one is the most effective one at any given moment in time, didn't it?

31:42Anders R. Lange: So what that's what you suggested to you is don't tie yourself into one thing for five years. And I think that that's not only AI, I think it comes with software where I think there's been a lot of benefits in putting your spend for a number of years because you get better discounts. And that's surely still true. But what you're buying now could very well be, not obsolete necessarily, but a lot less valuable just in a month or a few months from now. So I think really thinking as a procurement organization, how long do you want to commit to this technology and how much do you want to put all your eggs in one basket versus trying to promote alternatives?

32:25Anders R. Lange: And that's nothing new in procurement. We try to say that at every junction we have, but when it comes to this technology and how expensive and consuming it can be, I think that's the most important thing that we have to stress with our stakeholders and with our own development of solutions. Be careful how long you tie yourself into something because you might regret it. Yeah, there's a lot of value in optionality right now. Just a follow-up question on relationships because you mentioned it at the beginning of the conversation as well. Do you believe that relationships become more important in the days of AI rather than less important?

33:04Anders R. Lange: I have always thought and banged the drum of procurement being a supply chain, being a relationship business. And I said that before getting this job in terms of building trust. And I think that it's as important as ever, but I think it's changed certainly because if you introduce AI and agents and a lot of the interactions between your stakeholders and your function can be in a chatbot or some type of technology. I think the fact that you then spend time and saying, how was this experience you're using? Did it get you the answers you needed? To, you know, what is your overall strategy and objective for this service line or for this area you cover?

33:51Anders R. Lange: Because if the same behaviors as getting involved late happens in AI, it doesn't make a difference for somebody to say, hey, I need the support with this contract that's coming due tomorrow. So still being able to be close with the business, I think, is as important as ever. But now there's the ability to do free range and do things through agents and to effectively be carried all the way to the finish line without human interaction. I think it's becoming easier and easier. So the fact that there's a human that can be trusted to help support that process, Because asking some questions that comes from experience or that weird context that the agent might not be trained on, I think is important.

34:39Anders R. Lange: And then just as a human being, if I were to say that I think the agents will take over and I'm studying at night for a different career, I don't think that would be as exciting either. So I think just as a human being, the fact that you can interact with people and solve problems with the help of AI, I just think is as key in procurement and supply chain as it is in many other functions. Yeah, one of the things I wrote down there, which I kind of took away from that, and I hadn't necessarily thought about this before, is that with all those interactions being agent driven, there are less touch points between humans.

35:14And so when you have a touch point, you need it to matter. You know, it's like each touchpoint is more impactful because you need to give a reason why somebody will want another touchpoint with you rather than take the quote unquote easy option of interacting with a chatbot instead.

35:32Anders R. Lange: No, absolutely. And listen, I think agents will become smarter and learn from its actions if everything goes according to plan, how you perceive it to be. but there's also corrections and we've seen being able to ask questions and be critical is important. I think everyone who's listening in here, and I'm sure yourself too, and certainly concerning myself, whether you do it with your kids or you do it for yourself, when the agent tells you or whatever AI you use tells you something and then you figure out, oh, that was actually not true. Just recently I went on vacation and it suggested I should go somewhere on a Sunday only to find out the day before that it was closed on Sunday.

36:13Anders R. Lange: That would have been nice to know. Trust but verify. Trust but verify. And I think that goes a little other. And certainly because of the speed of how AI and AI agents can execute, that trust but verify and making sure you have a human loop to do so, I think it becomes more critical, but also quickly somewhere where we can lose ourselves and not pick that up. Yeah, so my last question today, and this is probably always the hardest because it requires a little bit of crystal ball gazing, is what do you think is going to be true about procurement in three years that's not actually obvious today?

36:54Anders R. Lange: I think, I mean, I alluded to it a little bit before, I think our time horizon will be shorter because of commitments for three years. So I think that's kind of been touched on. I also think for me, this is the most exciting time to be part of procurement because we can solve things ourselves, we can create agents ourselves. So that piece where procurement is dependent on stakeholder business to figure out what to work on or IT and how to implement things, those things, I think, becomes a lot less of a barrier. that we still have to go out and talk to people like we discussed before in building relationships.

37:34Anders R. Lange: But I think that our ability to do that more rather than having to deal with back-end issues and bureaucracy and whatnot, I think it just frees us up to talk to people more. And if we can take that opportunity to address the business more, I think really a lot of the ways we become more relationship managers in the true sense of that. And then we use agents to facilitate what we're doing. So I hope that's where we're going. And I think just the pace of AI and certainly in our market, I think it'd just be exciting to see the things we can't envision right now that will happen. I also think there's so much saturation in the market when it comes to solutions and to a finite list of problems.

38:21Anders R. Lange: I think there probably also have to be a little bit of taking a deep breath and see how are these solutions and how we go about things. Is it actually solving the problems we need to do or is it again just amplifying the same mistakes that we've done in a non-AI world or the same challenge we have in a non-AI world. So I think that there'll be a little bit of a wrecking as to some of the same core things as a human and as a procurement professional will still be important. I think that goes back to relationships. Yeah. Well, I want to thank you so much, Anders. I think, you know, maybe I'm biased, but I'm very excited about the future of procurement where before I may have been looking at it thinking, you know what, it's a little bit samey every year.

39:06You know, we kind of operate within the same box where we have our traditional processes and we're trying to get a little bit faster with our traditional processes and maybe a little bit smarter in the results. and we're trying to increase our influence within the organization that ever changes. But now the whole world is kind of a lot more open to creativity. And the government never used to be, I don't think, a function where creativity was rewarded because it was very process-driven. I think that's what was really exciting now is that we can now kind of re-envisage the role that we can play and actually we can make it happen now.

39:46We're not constrained by the tools that we have access to.

39:48Anders R. Lange: I agree. And if I tell my peers anything, if we can, while focusing all our time and talking about AI about token consumption and data management, which is certainly an important piece in figuring out how you allow yourself to have contracts that doesn't just consume money every time you spend it. But that creativity in solving problems, I think, is excellent. Some of the examples we have on our AI agent rosters is exactly that. And I can't tell you how many conversations I have with C-Suites people in my organization and elsewhere about these things that we've never done before. So the excitement is real.

40:27Anders R. Lange: And again, procurement is still very much involved and responsible for bringing in new solutions. So not only, you know. You're doing it for yourself, but you're doing it for the rest of the organization too. So you get to see all those cool things that are happening. So just stay curious and use some of those other things you see as inspiration. because it's such a fun time to be in procurement. Well, it's a great place to leave it. So Anders, again, I want to thank you so much for joining me on the podcast today. If anyone's listening, wants to connect with you, we'll put a link to your LinkedIn profile in the show notes.

41:01So anybody can go to outofprocurement.com. You can search for the Eugenic Procurement Diaries and see this conversation with Anders. There'll be Anders' link to his LinkedIn profile there and a little bit more of a write-up on the conversation. So Anders, thanks a lot. I really appreciate your time today.

41:14Anders R. Lange: thanks a lot Philip it was a pleasure appreciate it thank you

41:24if this episode struck a chord with you please do send it to somebody we grow here at Art of Procurement through word of mouth and that would be really appreciated you can also support us by giving us a thumbs up a star rating or a review wherever you listen to your podcasts since 2015 we've built the world's largest free resource for procurement professionals looking to elevate their impact our resources span podcasts like this videos blog posts papers and events to join us on the inside and to ensure you never miss an episode a webinar event or a post please do subscribe to our weekly newsletter this week in procurement you can do that at artofprocurement.com slash subscribe that's artofprocurement.com slash subscribe thank you for joining us today and i I look forward to seeing you again soon.