886: The AI Performance Gap: What Procurement Leaders Are Doing Differently W/ Christopher Sawchuk

0:05Christopher Sawchuk: Hi there, my name is Philip Leidson. I want to welcome you to the Art of Procurement podcast, the podcast that helps you, a forward-thinking procurement professional, position yourself and your team to proactively take advantage of the revolution that's taking place in procurement today. By interviewing industry trailblazers and sharing insights from our own experiences, my team and I pull back the curtain and shine a light on the strategies, tactics, and tools that procurement teams are using to elevate their impact. And today on the show, I'm delighted to welcome back repeat guest and friend of the show, Chris Sawchuck.

0:40Chris is a principal and the global procurement advisory practice leader at the Hackett Group. I invited Chris back onto the show today to talk about the Hackett Group's latest research on the performance gaps in how procurement teams are using AI. So first of all, Chris, welcome back to the show. Thank you, Phil. I look forward.

1:00Now, you've obviously been benchmarking procurement performance at Hackett for a long time. We've talked about a number of the studies that you've done on the show. How does this moment in time kind of compare to earlier inflection points for procurement, such as the rise of category management or the digital transformation wave? Yeah, yeah. It's a great question. And your comment, yes, we have been benchmarking the procurement performance at Hackett for, you know, I think it's over 30 years now. You know, it's interesting because as you look at it over this longer period of time, you know, somebody who's been in this space for a bit of time, it's an opportunity to reflect on, you know, what you're calling some of these inflection points, you know, over this span of time.

1:46And, you know, I would say that, you know, one of the inflection points you highlight in category management is something that occurred, you know, it differs for different organizations. But if you went back, you know, even two decades ago, you started seeing the uprising and the, you know, this, this, this, you know, focus on category management becoming more strategic. Yeah. No, it was around 2001, 2002. I think I was involved in the first category management implementation. So it's been a while. Well, I always saw it because people would talk and use the term category management. But then when you dug into the covers and you started looking at what they were doing, I would argue that it was strategic sourcing.

2:28And in some cases, being strategic was probably pushing it a little too far. It was just sourcing. So I think we started to reflect on just understanding what it was, but it didn't actually come to fruition for some period of time. And this was a major inflection point. But you also talked about digital transformation. And then what we're looking at today is this AI-enabled procurement. But if you reflect on this over this period of time, I mean, category management really was the opportunity for us as procurement in terms of changing the way that we actually made decisions around what we're going to go do, how we will intervene in a supplier market.

3:11And then what digital transformation, which we were talking about the time when category management came to fruition early in the 2000s, 2001, et cetera. When you look at digital transformation, that really started to arise and emerge in that 2017, 2018 timeframe. But what digital transformation did is really focus and improve how we actually did work, how we executed processing. Now we have AI sort of emerging a few years ago. And in many ways, it has the potential to redefine work as we do it today. So it's fundamentally different. And even how we make decisions. I like to say that category management changed our mindset around procurement, whereas digital procurement changed the tools that we utilize.

4:09And what AI has a potential of doing, I think it's more than a potential, I think it will, is really changing their operating model within procurement. Yeah, I see the operating model as the biggest opportunity. Technology being an enabler of operating model change, but it has to start in how we actually do procurement. And even what is procurement? You know, what are the boundaries around which we place when we think about what procurement does, which have all historically been pretty rigid. And now we can open our mind to think, OK, we have access to all the supply market. How can we be thinking about bringing that supply market into our organizations and supporting our organizations in different ways?

4:51We perhaps couldn't even think about before. Yeah. And it's interesting that you say is that, you know, how do we define the boundary of what procurement is and supply management? And we've always defined it as a set of processes. And recently, I wouldn't say recently, but a few years ago, we redefined what that is. What are those activities? What are some of the activities that we're doing today and we weren't doing that really wasn't defined within this set of processes and activities that we did before? But I think what we have is an opportunity to potentially redefine that even further, potentially broaden it even further as we look forward.

5:32Now, the research describes the performance gap, I think, between organizations that redesign procurement around AI, as we were just talking about, and those that simply automate what they already do. Can you talk about what that gap looks like on a day-to-day basis? Yeah. You know, anytime you look at gaps, gaps are typically between organizations who are less performing and those that are better performing. In many ways, the leaders versus the followers. And, you know, if you look at today from an AI lens, if you look at the organizations that are using AI, and I would argue that all organizations are using AI, the differences in terms of how are they doing that.

6:16And so followers, those that are creating this gap and are more lagging, and I would say they're lagging the leaders, they're using stuff more from a productivity standpoint. In many ways, they're improving things incrementally. They're using AI to summarize emails and draft emails. They're summarizing contracts, et cetera. But what's not changing is the processes and the activity and the way that work is being done is largely unchanged. Now, what we're seeing in the leaders, and again, this is a few, and again, this is an opportunity to see more and more. But what we're seeing is that they're actually looking at the opportunity to redesign workflows and how work is done through AI.

7:08they're changing you know as we were talking before they're operating models they're changing staffing models they're looking at different performance metrics within their organizations and so you know what you're seeing in these organizations category managers and we're talking about you know category management how that evolved back in 2001 but category managers are spending less time collecting data all of the work that we spent it's laborious i was out in on the West Coast this week, doing a session training around this. And they're spending more time and they had the ability to spend more time focusing on supply markets, the strategies, how we engage.

7:51And again, that's driving some significant gaps in performance. One of the things that I can reference is the paper that we put out recently around the impact of AI and that we're seeing another level of performance. There was a level of performance that digital got us to. And we've always used the term digital world-class. What we're seeing is a level of performance put on top of that. And this is those organizations that are saying, all right, we have the digital foundation within our organizations. We spent years putting that in place. And now what's the new performance level that's being set by what we're talking about with artificial intelligence.

8:39And so, you know, it is not really around the adoption of AI. Most organizations are adopting it. It's how they're redesigning their model that's really what's causing this gap to occur. I'm not sure if you have data or not, or whether this is more of an anecdotal question. but when you think about those organizations that are leading the charge, that are taking that step, that are rethinking kind of their operating model, and they are redesigning their operating model with AI in mind, is that 5 % of procurement teams? Is that 20 % of procurement? Like, what are we looking at on the continuum of, you know, how many as a cohort versus the rest of procurement kind of world is actually on that journey to to to become world-class or has already become world-class uh you know leading the way kind of thing you know i would say we're in the early innings um you know do we have organizations that are redesigning and they're taking that approach to redesign the operating model through the lens of ai the answer is yes are they there yet?

9:50Christopher Sawchuk: No. I don't know any organization that is fully there. I had the opportunity just even earlier today to speak with a leader of procurement, a large consumer products organization in Europe. And we were talking about the changes and we spent a lot of time talking about the operating model change that has to occur. But it's a cultural change. It's a mindset change. It's a shift in terms of how I think. Because I have to think about work, and I have to sort of step back and say, if I have AI, how do I use that to redesign work versus how do I embed AI into the way that I do things today? How effective have pilots and proofs of concepts?

10:43I feel like there's pilots and PLCs going on everywhere. And yet, you hear data, but there's not procurement specific around the successor or otherwise of proof of concepts and pilots, which I think is a little bit kind of hyperbole versus what we see within procurements. yeah but but do you see that the number of pilots and plcs that you see kind of do they lead to increased performance or is there still some kind of gap between being able to take a small use case testing on a small you know a small data set or a small activity or use case set actually translating into helping organizations on that journey to world class yeah one of the things that is becoming more apparent with many organizations is it's not, as I was saying before, it's not the adoption of AI that's the differentiator today.

11:40We're all doing it and you see differences. It's really the way that we're doing it. And I believe you have to go through various phases. If you look at over the last year, you could argue that many organizations and probably most organizations have been more focused on what I'll call experimentation. They're looking at AI more as a technology project versus what I'll call a business transformation initiative. In many ways, the pilots that were put in were isolated from the core procurement workflow. An example would be if you're focusing on CLM, the entire lifecycle management and the life of a contract.

12:25What you saw where organizations were doing pilots, hey, we're going to use AI to summarize a contract quickly. Yes, it had an incremental productivity benefit to do that. But what it didn't do is it didn't get embedded in that workflow. And it didn't fundamentally reshape how we think about managing the lifecycle of contracts within organizations. And so what you saw, and you still see this a bit today, but it is changing. The way success was being measured was based on the number of use cases that were being deployed rather than what I'll call the business outcomes that are being achieved. And so things have changed.

13:09And so I know that if we went back two years ago, some of the results that we were getting out of the models, we were suspicious. They weren't always accurate, but the models today are much more capable. You're seeing the quality of data, much more focused awareness of the importance of the quality of data. And I think the most important thing that has really transitioned that's going to allow this to accelerate going forward is that organizations are becoming much more comfortable with AI supported decisions. And so the first wave had to be that we had to understand and learn what AI could do.

13:50The second wave is really redesigning procurement around what we can or what it can do as a capability. Where do you see that trust? Because I think a lot of it is down to trust. When you say that we're thinking, we're kind of taking on board more or enabling AI to better support us in our decision making, and that comes down to trusting the outputs that AI drives. Where does that trust come from? Or how have you seen organizations build up that level of trust where they can now more comfortably move to semi-autonomous processes, for example, or your data that they don't necessarily see how that data has been derived, but they trust the output of that data so they can use it within their category strategies?

14:38Yeah, it's experience. It's engaging with that. You can't develop trust. It's like an individual. You have two people. You don't develop that trust until you develop a relationship with it or with that person. And that relationship is built on, you know, some fundamentals. Do you trust you? Is the person that you're engaging with somebody that, you know, that you enjoy? I mean, there's different things that you're using. They're reliable. They're reliable, right? Same thing with us in terms of working with AI. Just embedding it to an environment and just expecting that everybody's going to trust this and just go forward.

15:24That's not how we work. Our human nature is it takes time to build trust. And we had to go through that phase. And I think with more and more people using it and using it on a repetitive basis, that trust is continuing to grow. Yeah. I mean, two years ago, I had people would tell me or even a year ago, oh, I got it out of there. wasn't there's nonsense data, et cetera. And so you're hearing less of that today. But it's like anything, you hear stuff from people you know from a human. Do you trust everything those people tell you? Probably not. I mean, you're going to have to develop that no different than engaging with AI.

16:05So a very similar kind of relationship building, I'll say. it's interesting as well i think we have a higher standard on um what we will accept from ai than what we will accept from a human i've said that for a long time is that you know yes we we hold ai to a different standard than we do other individuals that we work with yeah um now the the report um uses purchase to pay as one of the clearest illustrations of the opportunity associated with AI. I wonder if you could walk us through kind of where that value actually comes from. Yeah. You know, it's always interesting. And anybody that knows me, no, I don't get that excited about purchase to pay.

16:47I very much like the strategic aspects of supply management, engaging the markets, the risk, et cetera. But when you think about this, you know, where does a lot of the exhaust, the data, the, you know, get created. And much of that is done in the transactional relationships that we have within the supply markets with our suppliers, et cetera. So it becomes this massive source of intelligence and data that we can access. Now, certainly AI can, you know, engage in, I'll call the downstream types of activities. We're going to be using it And hopefully we're going to get to the point where there are more and more, and I think this is a significant opportunity of freeing up that capacity.

17:37We've been fighting and wrangling with that for a long time. I think this is where we're going to see an acceleration here. But we're going to get benefits, less exceptions on invoices, less supplier inquiries. But all of that base that we're building is going to have an impact across the entire value chain upstream, you know, in the whole source to settle process, you know, better supplier data, which means better supplier intelligence, which informs better sourcing decisions, better supplier selection, etc. So you're going to see that. And, you know, the old idea is that, you know, the biggest AI value isn't going to become, and again, this is looking at it incrementally from just processing things faster.

18:25It's going to come from the decisions and making better decisions, you know, in some ways before that invoice even exists. Yeah. And that becomes a talent issue. And people probably listening to the podcast are tired of me talking about this. But the time to be thinking about what to do with that freed up capacity is before you freed up that capacity. because we have spent 20 odd years that i've been in procurement talking about this desire to be more strategic with nobody really knowing what that means and you actually alluded to it before in terms of category management well is that strategic sourcing well is that just regular sourcing another way to um uh you know to to have suppliers believe that we're their friends while we're just trying to use leverage on them right and and so i see a risk there that we we get these efficiencies, but we don't actually know what to do with that time when we get that time.

19:24And that feels to me like one of the biggest challenges, but probably opportunities that come along with the efficiencies that we can gain from AI right now. Yeah. You know, I think one of the ways you can look at that, you know, because when you look at the report that you highlighted, you know, we highlight some pretty significant or high percentages in terms of freeing up capacity. And I think not everybody does this, but what we're seeing in these organizations, these that are all called leading organizations, as we were talking about before, they're not framing this as a headcount conversation.

20:04They're framing it as a capacity conversation. And so the conversations that they're having, the discussions within their organizations, they're asking themselves, you know, well, with this, what work disappears? What higher value work becomes possible? And how do we redeploy the talent that we have today to engage in that? So, you know, the conversation shouldn't be about how many people we can remove. What the conversation needs to be is how much more value can we extract from that same team and how much more value can that team create yeah um yeah and i think the number was it 81 percent yeah in the report 81 reduction in staffing requirements i should say really staffing needs when you look at apples to apple comparison of processes that teams may do today versus teams that uh processes that teams may do tomorrow so that's essentially your excess capacity which is ready to deploy i don't feel how we respond to that is the the existential question for what procurement becomes yeah because if we if we don't respond to that in a way that redeploys and trains teams to do more high value added activities then that capacity just gets taken from the system um but if we do then that we're able to redeploy those resources to do things and create this virtuous circle that becomes far more impactful for the organizations that we serve.

21:45Yeah. So it's imperative that we have a place, a path for that capacity. Without that, yeah, you run the risk of headcount reductions and staffing reductions within your organization. Now, when we come to therefore designing the kind of future

22:03Christopher Sawchuk: state operating model you know those operating models um include what do the new rules look like uh you know what does governance look like what they do with that capacity uh that we talked about how specific do you think that leaders need to be in designing the future state while they're before they start on their um their journey versus experimenting seeing what works trying a few use cases and certainly finding efficiencies and then not necessarily sure where to go next. Right. I don't think you have to have every answer before you start. But what you do need, and this has been always the case, is you need to understand your destination.

22:49That has to be clear. It's your vision. Where are we going? I always say, if you don't know where you're going, any road will take you there. You have to have a clear destination, but you may not every answer as to how you're going to get there before you're starting. But what you need to do is you need to understand, you know, what work can AI perform for us? You know, what work will our human capital within the organization perform? We have to think about governance and how we're going to make decisions. To your point earlier, what skills are going to be needed, you know, going forward how will we measure success you know so it's not about buying technology and designing processes around that product it has to start with the outcomes the business outcomes that we want um this becomes an operating model decision you know technology has to support the operating model that we're designing the operating model can't be forced to fit the technology.

23:53How do you find that the procurement leaders that you talk to are dealing with kind of the ambiguity of not even necessarily knowing what the destination should be? Because I think that we're all on this journey of exploration right now, where we don't necessarily know where it's going to take us and what the capabilities are going to be like in six months time or 12 months time or whatever. So we're trying to make decisions when there's a lot of unknowns. How are the procurement leaders that you look at and admire for what they're doing within their teams, how are they kind of making those calculations or balancing the knowns versus the unknowns?

24:33Yeah, it's a great question. And I'm going to go back to our forum that we did with our clients back in May. And when we did that, one of the exercises that I did with the group, we had roughly 50 different companies within the room. And what we did is we, I shared insights in terms of what is the procurement operating model. You know, this includes our human capital slash talent. This includes our processes. This includes our technology. This includes our data and information that, you know, enables that over our operating model. It's our organizational structures, et cetera. And so what I did is I divided the room up into these different components.

25:14and we gave everybody about 25 minutes, maybe 30 minutes to think about. I didn't want to hear about where you are today. I didn't even want to understand the path of getting there. What I was asking for is what are the characters, what's it going to look like as we sit here three to five years out in the future? So after 30 minutes, I, you know, every table, you know, went through and they, they shared their insights as to what they came up with. And I would argue that when I look back at what came out of that, most of where we were and most of what I was hearing, I would argue it was maybe a year to a year and a half into the future.

25:57And I sat back. I said, why is that? Why are people struggling in putting together a vision of the future? And it really comes down to the speed and the uncertainty of how fast things are moving. And I use the example, if you went back a few years ago, maybe two years ago, we weren't talking about agents. You and I, we go to a lot of different events. That wasn't part of the conversation. Today, it's a significant piece of that conversation. And when you look five years from now, what technology is in our blind spots that we don't know that have a fundamental impact in terms of what we do? I think that's what's happening.

26:43It was much easier to do this 10 years ago. An organization could sit down and they could put a vision together because the velocity of change on the outside wasn't as great as what we're seeing today. Yeah, we'd have our five-year plans, our five-year transformation roadmaps. that all kind of looked the same as each other's, just depending on where an organization was on the maturity curve. And you look back and think, just the last two or three years, as you said, we've had the first wave of AI was, our consumer kind of grade AI that we're all able to wrap our heads around was chat. Then the next one is agentic.

27:23And then what's the next one? The next one may be us all having virtual teammates, you know, chief of staffs in our pocket that's going off and doing things proactively and reporting to us. And yeah, I'm sure that's not that far away. And then what comes next? What's the impact of quantum, right? In terms of how we do things. And this is already being explored by organizations. But yeah, I think this is where some of those challenges are. Do you sense excitement around those challenges or trepidation? Depends on the organization. You know, I had a conversation with a CPO recently and, you know, one of the things is, is how do you ready the organization for these fundamental changes?

28:08And his feedback was, you know, we couldn't move forward as you start looking at all of the things that are happening in the marketplace. I mean, if you're in manufacturing, you have a physical goods supply chain. You look at the changes that came out of COVID disruptions and how that fundamentally changed what we were doing in procurement. And you start sitting there and say, well, how do we ready ourselves now for AI going forward? And it fundamentally requires a cultural, a mind shift change within the organization. If you don't do that, how you can progress going forward? And so that's fundamental, you know, that we have to go do.

28:45We have to ensure that we engender a curiosity. We have to engender a model where we challenge the way that we're doing things today. We truly rethink. And that's different than the incremental improvements that we're seeing so many organizations doing today when there's this bigger opportunity. And I think that's what's holding it back. It's a fundamental mind shift that has to occur within these organizations. Yeah. And that's, that's not, it's the, the pace of change is driven by that rather than the pace of you know, the ability of the technology to support it. Yes. Yes. Yes. It'll be, all of these things we're talking about, the technology is moving rapidly and the capabilities are there and it will continue to enhance the friction is how do we embed that?

29:43How do we really rethink, reimagine, transform our operating models to fundamentally do things differently to create better business outcomes for the enterprises that we support? Well, I have one last question just to wrap up. And that's, you know, I'm sure there is CPOs listening today who have done, as you said, you know, kind of the beginning of the journey of AI. So they are using AI within their teams. They may be using it to do certain use cases, but they're under pressure to do more. They're under pressure to be further along that journey to become a leader in the use of AI within their procurement organizations, even though they've probably got the same challenges that you talked around there around mindset, capability and all those things.

30:26From what you have seen from all of the teams that you work with, I know this is very circumstantial, so it's hard to give me one thing. But is there kind of one thing that you would encourage them to really think about and focus on in the next 90 days to continue to kind of take their team up that maturity journey when it comes to the impact and the outcomes that have been driven by AI? I think most organizations, what they need to do is do what I call AI opportunity assessment against their entire process, the procurement processes. You know, ask themselves, where is the work today? Where are the bottlenecks?

31:04Where are decisions being made? You know, what activities create the most value? What activities are you doing today that consume the most capacity? um you know a lot of times organizations we saw that through digital transformation start with tools what the leaders do is they start with work redesign and so you have to ask yourself you know you know let's not start by asking you know where can we deploy ai that's what we've been doing with the experimentation the use cases etc but what we now have to do is shift that we have to shift to asking you know about procurement work and and and how will it operate if ai was available and embedded in it well chris i want to thank you so much as always for joining us i think it's really important that um you know in in this day and age when uh there's a lot of unknown that we share in kind of well things organizations are using ai in certain ways and these are being successful in doing so, that it's not, you know, it's more than just hype, you know, as you go on that journey.

32:14And if you are coming up against challenges, you know, you work through those challenges because there is something brighter on the other side when you kind of start to really deploy AI across your organization. So thank you as always. What we'll do is we'll put a link to your LinkedIn profile in the show notes for anyone who'd like to reach out as well as to the Hackey groups website um and those will be on our show notes page so anyone listening that's art of procurement.com slash podcast that's art of procurement.com sorry i can't even say my own website name art of procurement.com slash podcast it's been a long day i think um chris one last time thank you so much for joining me today if this episode struck a chord with you please do

32:58Christopher Sawchuk: 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 look forward to seeing you again soon