Voices on Planning: The AI-powered future of supply chain

0:09Daphne H. L. H. Wong: Welcome to McKinsey Operations Voices On series. This is where we dig deeper into the detail. We bring into sharper focus some of the tactical operational challenges and look at how operational excellence combined with tech and AI has the power to change the game. Today we're talking about Advanced Planning Systems, or APS, and the AI-driven transformation APS leaders must now put to work. It seems like each decade, supply chain and APS leaders come face to face with a technological shift that forces them to completely rethink how they work. First, it was moving from spreadsheets and legacy systems to enterprise APS.

0:45Daphne H. L. H. Wong: Then it was the shift from on-premise software to software as a service or SaaS platforms. Today, it's the move from software-driven planning to AI-enabled decision-making. The ones who are leading here aren't necessarily the businesses that have the best AI tools. It'll be the organizations who are willing to challenge assumptions, retrain talent, redesign processes, and let go of legacy ways of working. Joining me today are Andrew Shi, Principal Product Manager at McKinsey's Quantum Black AI, and Mike Raftery, Senior Global Asset Lead for Digital Supply Chain here at McKinsey. Mike also co-founded and then became the CEO of SEM Connections, which McKinsey acquired in 2022.

1:28Daphne H. L. H. Wong: Andrew and Mike, welcome. Well, thanks for having us. Glad to be here. Mike, you've spent 25 years in supply chain focused roles and then you led SEM Connections for several years. No doubt you've seen the industry change over time. Take us through some of those changes. Yeah, I mean, I'm kind of dating myself, but I think when I reflect on it, this is probably the third sort of extinction level event in the industry that's happened to force change and evolution. I started my career in the early 2000s when planning software was just getting started. And from then it was sort of growing on MRP, getting to S &OP processes and really trying to just figure out how to get an end-to-end solution going.

2:05I think about 10, 15 years ago, something like that, there's a move to SaaS-based platforms really driven around simulations and IBP processes that changed the game for the last generation. And then right now, just in probably the last six months, maybe less than that, we've seen a whole lot of change with these AI tools coming in where it's kind of getting through some of the hype and understanding what really is going to emerge from the ooze, so to speak, and the evolution of it. And that's where we're at today. And it's an exciting spot to be. It keeps me engaged even after all these units.

2:39Daphne H. L. H. Wong: Never a boring moment, I imagine. So, Mike, what did the APS planning system's users need to unlearn as they transitioned from on-prem to moving to a SaaS model? Yeah, this was a huge change for the first, I guess, 1.0 to 2.0 of this generation of platforms. I remember being in a lot of meetings with senior executives. First year it came out just saying we were never moving to the cloud. Like, never. And then a year later, nobody cared. It was just how it is. And they kind of took it, got through the challenges, got through all of the concerns, governance. That was a huge change there was how do we take it from, you know, something that we have a server to manage to something that's cloud managed.

3:20And right now, that evolution is, I'm feeling that same vibe, that same like, well, we're just never going to do that, or we're never giving our data to AI, right? And it's the same type of mentality. And I think people are working through the governance of that. What does it mean to expose your data? What are we comfortable with? What policies need to exist to do that? And it's a lot of those transitions that have a similar flavor, but in a much different, more accelerated, more exciting manner with the move to AI.

3:47Daphne H. L. H. Wong: And Andrew, let me bring you in here because you've got a lot of experience now in working with teams who are moving from SaaS to the AI transition. What are you seeing and how is this shift different? I think the biggest shift is really just how the cadence almost evaporates entirely. Whereas before, teams and organizations plan on daily, weekly, monthly cadence. And now sort of with the genetic AI, you can have agents watching data and orchestrating constantly. You really can plan when reality changes, not when the calendar shifts, if you will. A big shift is about sort of accommodating that, but then also on one hand, setting up your organization, your data, your operating model to get the most out of it.

4:24And then, like I just said, almost such a fundamental shift demands you change your operating model pretty fundamentally as well. So that's probably the big shift I'd call out, the always on aspect of it.

4:33Daphne H. L. H. Wong: Andrew, as you think about that, how does the role of the planner change in that? What new skills become much more valuable now? Fundamentally, one big one is almost teaching in the sense that when you're setting up agents to sort of watch and orchestrate your cadence for you or you're planning for you in real time, encoding the context of your knowledge in such a way that these agents can be trusted to do so. That's one. The second would be deciding and maybe deciding to almost understate the importance of it. But human loop intervention will still be super key here where this can't be a fully automated process.

5:06You still need that human accountability, those decision points, that expertise in the moment to make those decisions. So being able to make those decisions in the moment, that's probably the second big one. The third one is really maybe shifting from on-the-ground execution to being more of an orchestrating role. So sort of tying the previous points together. Instead of being the one doing work day-to-day all the time, maybe more the time you're setting the goals, setting up the guardrails, and actually deciding what's automated versus what's done manually. I'd call it those three big shifts.

5:32Daphne H. L. H. Wong: So would you say that planners now will spend less time actually planning and more time governing the AI? Yeah, yeah. Governing, teaching, coaching and intervening, all those things. Interesting. Mike, let me bring you back in here. So, you know, if that is kind of the backdrop, how do planning organizations prepare for AI? What should they prioritize as they start to think about their reskilling as well? But Mike, first to you. I think there's a lot of concern and maybe anxiety around that. Do we even have the right team to do it? And I think there will be some evolution and upskilling on it.

6:06I think there's two parts to it in terms of how does the organization prepare. There's going to be a lot of questions around governance models, architecture models. How do we actually stand this up? And what does an enterprise-grade AI solution require from us organizationally? That's probably one area that needs to start having discussions right away. As far as the planners go, I've heard a lot of different concerns on either workforce reduction, and I feel like that's not at all the case. To Andrew's previous point, get out of that cadence, get out of that mindset, which is really hard to do if you have planners that have been doing, you know, the weekend job runs because it took a weekend to run.

6:43Well, what if it doesn't? You know, why do we have this meeting? Or why don't we have this meeting? What would you do with this information? You know, what questions would you love to answer if there was infinite amount of analysis and resources to do it? Those are the type of questions that when you start using these AI embedded tools, really take a pause and a mind shift set to say, what can we do here that we couldn't do before? Because if you assume infinite amounts of scenarios and simulations, your questions become a lot different. And so planning becomes less of a cadence to say, well, we have a Monday morning meeting because this, that, and the other thing.

7:19You may just end up saying, well, you know, when these occur, this is who I call and this is the information they expect. And eventually that will just become planning. So it's a huge difference in what that task looks like. And I think it'll evolve over time as some of these innovations come to the market.

7:33Daphne H. L. H. Wong: And Andrew, let me bring you in here. So let's talk a bit about, you know, how should planners consciously think about reskilling? What are some of the reskilling priorities they should be thinking about? Yeah, I'd say both planners and even planning organizations, probably the biggest shift. I love Mike's point earlier about sort of how these changes will happen. From my perspective, I almost frame it as, what are the capabilities needed to enable all these shifts, right? So probably the biggest one for me is, and this is maybe a reflective of a broader shift in, you know, agentic AI as a whole, but making sure the context is set up, making sure that when we talk about AI or agents sort of orchestrating and enabling all these, that they are able to understand the data and the scenario well enough that they can do so reliably.

8:12We've all heard about the risk of hallucinations in AI, sort of how that can go comically or horrifically wrong, depending on the situation. So we're setting up almost the semantic layer, which is kind of one of the things we call it, so that agents have the context, so they can actually understand the data and the situation well enough to run reliably, basically. That's probably the biggest one. The second one is maybe related, actually. Sort of taking a holistic, you know, some people call it a platform approach, others call it an operating system approach, but really it's just building that orchestration layer so that you're not building sort of siloed point solutions for, you know, use case by use case or, you know, widget by widget, but taking a holistic approach to if you're going to rethink how you work, what's the overall approach, both technically and operating model wise, that'll enable all this.

8:55Daphne H. L. H. Wong: In some sense, actually liberating. So, Mike, if I bring you back in here, let's take a clean sheet approach to this. I mean, if you were designing a planning organization from scratch today, what would you not build into it? I think a lot of the core skills are there. I think the danger that you would have is you want people, to Andrew's point, that are working a lot with these models and know how to manage them. And so I think what you would want to have are people that can actually challenge effectively. One of the dangers in this hallucinations is that these models come back sounding so confident.

9:25Like, yes, this is the thing to do. But if you don't have that governance and expertise to say, wait a minute, that doesn't sound right, challenge the model. If you get complacent, I think that's probably the biggest danger that organizations would have. And so you really need to have these individuals be critical thinkers and problem solvers. It needs to truly be somebody that can see complex problems in real time and understand the implications of it. So the skills are, I mean, those were always the best planners, but they weren't always the planners, right? These were the people doing sort of network studies.

9:54And I think that's really where, you know, you want to lean in and reward those kind of behaviors.

10:00Daphne H. L. H. Wong: So people who are consciously challenging the whole time. Yeah, the difficulty is going to be in that, though, is where does that next generation of expertise come from? Generally, those that have that experience got it because they were on the phone getting yelled at by a Walmart or a Target, right? They couldn't find order on time. And so they kind of built this experience over time. I think the challenge for organizations is how do you mentor the next generation so that this talent doesn't just walk out the door and you're reliant overly much on the AI models? That's probably going to be a challenge for a lot of different groups within an organization.

10:34But to me, that's really the risk in making this sustainable.

10:37Daphne H. L. H. Wong: Interesting conundrum. But I assume that some companies are actually doing this well and are really on a good road. So what are the characteristics that have helped them to find that success? You know, what we see a lot of companies doing right now are sort of putting their toes in the water with proof of concepts and pilots and almost too cautious, I would say. But I think it's also the right approach. There's a lot of new skills, a lot of new organizations. I know, Andrew, you've dealt with a lot of companies that are dealing with questions around how do we make what is a pretty cool tool? Like, you know, you can go and download Claude or whatever and Vibe code an app in an afternoon.

11:11I think they're still dealing with what does that mean from an enterprise level? Yeah. And honestly, that's a big part of it. You know, I've seen a lot of organizations that they sort of let a thousand flowers bloom, right? And everyone builds their own sort of little POC or little widget, little application. And what they end up with a few months is, oh, you've got a bunch of cool applications. or is the impact? And that's when the awkward silence sort of kicks in, right? So I'd say one of the key things, and there's a whole bunch of other points, but the main one is to actually close the leap and measure it.

11:36You want to see what you're building is actually leading to impact, whether that's, you know, financial or operating model-wise or something else. But you want to know that what we've done has moved the needle in the way that we expected it to. A lot of what's old is new again, right? It's not necessarily getting the coolest app or, you know, vibe coding, the coolest little widget. I mean, that's all fine and well and good. But the ones that are really having impact are doing the hard work and the heavy lifting to understand how to make these things enterprise grade. The risk is that you end up supercharging your organization with a thousand spreadsheets everywhere with these vibe-coded apps that are unsupported.

12:08Data governance becomes risky. Data security becomes even riskier, right? That's really the downside of that. That's the cautionary tale is if you let everybody have a, you know, a cloud code license and go nuts, well, now you have lost control of your organization's data, decision-making, governance, security, right? We've had some clients that had experience with that and they pulled it back pretty quickly. So I think it is taking a, I think conservative is maybe has a negative connotation, but a purposeful approach. We use that to say, you know, how do we actually want this to run an organization, not just a cool party trick?

12:42And those are, I think, nuanced decision points, but the ones that will be successful in the long-term are doing things like measuring impact, measuring for goal. What are we actually trying to accomplish here? If you're just rebuilding your old demand planning solution in AI, you've missed the boat. What is the actual change that is going to move the needle in efficiency and impact? That's the questions they need to be asking, not just can we do it in AI.

13:05Daphne H. L. H. Wong: So if you were advising a planning leader who's going to be listening to this program, what would you encourage them to start doing tomorrow? Mike, let me come to you first with that question. I think it is understanding what is required in your organization's roadmap to leverage the AI capabilities without exposing your company to data risk or governance. This all works great until it doesn't. You know, we're kind of at in the hype curve things, getting into the part of some disillusionment as people realize that these are still some hard things. This is complex data. These are tough decisions.

13:36This is going to take some governance and control. Problem is though individuals are used to having Claude or ChatGPT at their home and think they can run and go with it. So there's a bit of entitlement that needs to be kind of reeled back in. So in my mind, there's two items here. One is culturally, help people understand the risk of what they're doing is or could be. And then the second one is more from an enterprise side. What do we need to invest in? What decisions do we make? Do we take it in-house or do we outsource these capabilities for governance, security, data storage, etc.? Those are probably the two most pressing questions I would be working on today.

14:12Daphne H. L. H. Wong: And Andrew, from your perspective, would you add anything to that list? Well, maybe not the list itself, because I fully agree with everything you said, Mike, but I'd almost layer sort of a lens on top of it, which is all those things I would advise a leader to really pick one tough problem in your organization, whether it's a particular, you know, a domain or something else to pick that and start with it. Start with that first. So I would say it's bigger than a pilot, but smaller than a full organizational transformation. The former isn't always guaranteed to be impactful. The latter could take years, but really pick one tough, meaningful problem you can start with, solve that, kind of go through all the things you mentioned, Mike, and then grow that accordingly afterwards.

14:48Daphne H. L. H. Wong: Great advice. So I wanted to ask a question of both of you, actually. Let me start, Andrew, with you. And help me finish this sentence. So if the SaaS revolution changed where planning software lives or lived, what will the AI revolution change? It would change what it really means to plan at all, if that makes sense. Because you're going from, I guess if I sum up everything we discussed up until now, the shift is almost from you're producing and executing the plan to more architecting and then governing the system that plans continuously while still preserving expertise. So it's a pretty fundamental shift.

15:22Daphne H. L. H. Wong: And Mike, how would you describe the AI revolution and what is going to change? I think it changes what's possible. And I know that's sort of a vague answer, but I'll clarify a little bit. I think planners are going to understand when systems are no longer the constraint, hardware performance, data models are less of a constraint. What questions can I ask that I never could ask before? And that's going to be an incredible unlock for organizations to realize. I think once that clicks, and it will click, the revolution will sort of come quickly as people understand how these get used in their daily jobs.

15:55I think we're still trying to figure all that out. It's a very exciting time for the possibility of what's happened in probably the last six to nine months in this space. And it's going to change even more quickly probably in the last part of the year.

16:07Daphne H. L. H. Wong: And I know we're never allowed to make predictions, but Mike, how long do you think that will take? Sooner than expected. I think it's Gardner that has that IT software hype curve, right? We're just coming off the peak of expectations into the valley of disillusionment. I think that's going to hit pretty quick. But because organizations like ours are trying to address this, we're asking the questions. We've been through these revolutions before. It'll happen a lot faster than usual. We're seeing clients understand the challenge in front of them and tackle it very purposefully, to use that word again.

16:39So I would expect that by early in 27, maybe, you would see a lot less piloting and a lot more utilization. You'll see some real strong success stories by next summer. Andrew, you agree? I agree, yeah. I think, you know, the token maxing trend is just catching on or rather dying out. And so I think people are getting a lot more serious and disciplined about it. So time frame sounds about right to me. I think if I have one thing to add, that there is an environment that is still changing underneath us. So, you know, I'm going to put an asterisk on all of these because there are all these third parties that are out there.

17:13And as new models come out and new, you know, new capabilities evolve, so will these questions evolve. You can't set and forget this has to be a constant change and a constant agile evolution to take advantage of it. Because it's not just, you know, the development cycles are so much quicker than they've ever been. The organizations will have to really stay attuned to this and live in that space. It'll be interesting to see how the ecosystem evolves in that. We're used to yearly conferences where everybody gets together and kind of sees what the new thing is about. That's way too slow. I mean, these things are happening monthly, almost weekly at this point.

17:45So how does the overall ecosystem, you know, incorporate new players like, you know, Anthropic, ChatGPT, all these tools that you have available to you are going to just be, it's a new player, a new player in the group.

17:57Daphne H. L. H. Wong: So I love what you talked about there, Mike, which is, you know, the curve moving into a valley right now, you know, where there's quite a lot of disillusionment, but we're going towards an upward trajectory. What, of course, is also going to be impacting us is the ecosystem of technology and the constant shifting sands around the technology, the solutions and the other tools that will be available as we move on that journey. Today's conversation has been about AI as much as it's been about the importance of reinvention. APS leaders have lived through one major transformation, from on-premise software to software as a service or SaaS.

18:35Daphne H. L. H. Wong: The AI era presents a new challenge, not just changing the technology stack, but rethinking how decisions get made, how teams collaborate, and where humans create the most value. The question isn't whether AI will change planning. The question is how we will change it. I want to thank my guests, Mike and Andrew, for this wonderful, thought-provoking discussion. And it sounds like we're going to have to be inviting you back in about six months to talk a bit more about how the sands have shifted. Andrew, thank you for joining us. Thanks for having me. And Mike, thanks for being with us. My pleasure.

19:12Daphne H. L. H. Wong: You've been listening to McKinsey Operations Voices On series, the place where we talk about operational excellence and how it has the potential to change the game.