Most organizations are measuring AI value the wrong way. While efficiency gains and productivity improvements matter, the leaders who are winning with AI have moved far beyond that conversation and backed it up with action.

In this episode of CGI’s From AI to ROI Podcast, CGI's Christina Fung, Helena Jochberger, and Andrew Donaher cut through the noise to share what's actually driving competitive advantage in the age of AI, drawing on frontline experience across manufacturing, government, healthcare and financial services.

Learn why so many organizations remain stuck in proof-of-concept purgatory, why efficiency gains alone are not enough to measure AI success, and what leaders must do differently to close the value realization gap. For additional episodes, subscribe directly on Spotify or Apple Podcasts, or visit the For AI to ROI Podcast homepage.

Key takeaways from the episode:

  1. Winning organizations treat AI as a modernization play, not an innovation one.

Leaders and organizations that are winning with AI are not just building chatbots or running hackathons. They are using AI to modernize legacy assets, change unit economics and deliver outcomes that were previously unattainable. One Western Canadian government client went from a year and a half of stalled analysis to code complete in two months and in production in four, using AI to solve a modernization challenge that had defeated multiple previous attempts.

“I think the ones that are really winning are the ones that are treating AI not as an innovation game, but as a modernization game.” – Andrew Donaher

  1. Efficiency is the starting point, not the ultimate destination.

Executives at more mature organizations have largely moved past conversations about speed. The real measure of AI value is its impact on core business metrics such as revenue growth, margin, customer retention and decision quality. Equally important are the harder-to-quantify gains such as removal of non-value-add work and the ability for people to find more purpose and meaning in their roles, and the conditions needed to create those gains. These benefits may not appear immediately on the balance sheet, but they show up decisively over time.

"We've actually stopped talking with our clients about speed. The metrics haven't changed, whether you're looking at your COGS, your SG&A, your revenue per FTE. We've gotten through that hype cycle and we're focused back on those metrics. And we're seeing those metrics change faster and that's the critical thing." – Andrew Donaher

  1. AI helps experts find efficiencies and areas where they can add more value in their work, but doesn’t make laypeople experts.

One of the most persistent fears around AI is that it will level the playing field in ways that undermine expertise or eliminate the need for it altogether. The reality is the opposite. AI amplifies what skilled professionals can do, freeing them from non-value-add tasks so they can focus on the work that requires judgment, experience and domain knowledge. It raises the ceiling of possibility for experts, it does not replace the expertise itself.

“It's making the experts more efficient. It's not making laypeople experts. If you put me in a Ferrari and you put Michael Schumacher in a Ferrari, they're very different experiences.” – Andrew Donaher

  1. The value realization gap is real and there are concrete ways to close it.

Many organizations remain stuck between two stages: moving beyond AI POC activity metrics toward genuine business and financial outcomes. Closing that gap requires treating AI not as a standalone technology initiative but as an integrated part of how the organization makes decisions, redesigns processes and builds operating models. The organizations that are furthest ahead are asking not where they can use AI, but which high-impact decisions they need to make better, faster and more consistently.

“Where are our most important and repeated decisions that we need to take and how do we make them better, faster, more scalable, more acceptable by the users?” – Helena Jochberger

  1. Do not underestimate the importance of organizational change management.

Technology is often just a small part of the AI value equation. Employees need to understand how their roles are evolving, leaders need to model the mindset shifts they are asking of their teams, and processes need to be redesigned (not just automated) to capture the full value AI can deliver. AI that is layered onto broken or outdated processes will simply produce amplified versions of the same problems, increasing stress and inefficiencies rather than enabling people to do better work.

"If we implement an AI algorithm and we don't do a proper process redesign, the adaptability of the people might not be there. Organizational change management is something 'oldie but goldie' which we should not forget when talking about AI and AI value." — Helena Jochberger

  1. Leadership sets the conditions for trust.

Leaders who model an agility mindset, communicate honestly about how roles and structures will evolve, and demonstrate the augmentation value of AI —while backing their words with action—build the trust that creates the conditions for transformation.

“The best leaders aren't pretending to have all the answers; we're sitting all in that experimental boat called AI. But at least they are clear about what they know and what they don't and what's likely to change. So I think walking the talk and being also role models in that kind of agility mindset is crucial for success. So that would be my first recommendation, clarity over certainty.” – Helena Jochberger

“The currency of the future used to be scale and in the future it's going to be courage.” – Andrew Donaher

Learn more and subscribe

Explore more episodes of From AI to ROI and learn how AI is transforming enterprises and government organizations. Visit CGI’s main AI page for insights, resources and updates.

Read the transcript

Introduction

Christina Fung (00:01)

Welcome everyone to the podcast today. We are going to go deep into a very popular topic from AI to ROI and how to get real true AI value beyond the hype. We hope that through this podcast, it helps you to understand how to unlock the real value in AI for your enterprise and for your jobs. I'm Christina Fung, Senior Vice President Consulting in CGI, leading global AI initiatives in the CTO group.

I'm excited to host this podcast today. And along with me, I have two great CGI leaders whom I worked with for many years. Andy, Helena, it's great to have both of you with us today. Can you introduce yourself, your roles and how's it like each day when you spend your time in CGI with the clients? And Helena, do you want to go first?

Helena Jochberger (00:53)

Sure, absolutely, Christina. Thank you so much. I'm pleased to be with you today. My name is Helena Jochberger. I'm the global vice president for both manufacturing and business consulting. So pleased to be here with you. And yeah, when you ask me, Christina, what is it like to be in these roles, discussing a lot with clients, discussing macro trends, discussing business and IT trends, and then of course, trying to find the best solution for our worldwide clients.

Andrew Donaher (01:26)

And my name's Andrew Donaher, I'm the Vice President of Consulting Services leading the Vancouver market. I'm the former Vice President of AI for CGI Canada and work with Christina and Helena as member of the global AI leadership team. Right now I spend, instead of my time on a plane flying around with you two, I spend my time helping Vancouver and the greater mainland clients, applying AI to make sure they're deriving value from their various projects and assets. Very glad to be here again with you both. Thank you.

The real value of AI in enterprises

Christina Fung (02:01)

So exciting, hearing all these things and going to hear more in the coming hour. So let's dive into it. We know that there is a lot of noise these days. We have read and heard many people talking about this AI everywhere. They use it for everything. But at the same time, we also hear a lot of companies and our clients talking about they don't really see the impact coming out, so they don't see and get the value at the end of the day. So what is really happening behind the scenes and where is the real value of AI in your opinion? So Helena, do you want to share some of your insights and what you heard from the client?

Helena Jochberger (02:45)

Yeah, absolutely. I'm pleased to do that because I think when discussing about all of the fancy AI technologies that are out in the market and that are really exponentially growing day by day, so that we, even coming from the field, can even not catch up with what comes out day over day, I think we must not forget that implementing AI properly to drive success is, first of all, a team effort, also on the C-suite, on the CXO part. So it's not a one-time effort of an IT responsible or of an operational responsible, but it's really a team sport.

And second, that we need to treat it a little bit more comprehensively. Since regularly the models, the classical traditional AI models are out in the field for years. For example, in manufacturing, this is not a new discipline. We were working with machine learning with pattern recognition since years and the term industry 4.0 at the time in 2010 was coined. So it's not really a new topic. What is new though is that we need to try to perceive the things a little bit more holistically. So meaning it's the data, it's the organization or the people, people make it happen. It's the technology of course and bringing that into a coherent state makes the success happen.

And what we see in the market furthermore is that at the very moment we are really in a proof of concept purgatory to talk with the words of Dante Alighieri, meaning we have great proof-of-concepts (POCs) created over the last 16, 18 years. We have great MVPs, but do we have cohesion, data continuity, AI continuity? Not yet.

Andrew Donaher (04:30)

Yeah, if I may just compliment that a little bit. So organizations that are just using AI for chat bots are the ones that are not winning, right? And so kind of to Helena's point, whether you're in POC purgatory, I think the ones that are really winning are the ones that are treating AI not as an innovation game, but as a modernization game.

The ones that are using it, it's changing the unit economics, and that's what people are understanding. It's not, for most organizations, if you're a traditional product company or you're a resources company, you're not going to start building your own LLMs and chatbots aren't going to solve world peace for you.

And so, by understanding how you're changing the unit economics associated with this and how you're understanding your delivery cadence for various projects. I think that's the real opportunity and the ability to do things that we've never done before. We're working with certain clients around modernization of older assets and things that took years or were simply not achievable now, we're able to do them. And I think that's the big difference that we're seeing moving forward, the currency of the future used to be scale and in the future it's going to be courage. That's what it's going to be.

Evolving executive perceptions of AI value

Christina Fung (05:52)

And when you're talking to clients and these executives, does AI value mean something different to them? I know a lot of times people started off around efficiency, right? Saying that I want to do things faster. And, but that perception of value probably evolves by now. And when you work with them, what do they say? Like, can you reframe some of this feedback to us around what are these values in their mind?

Helena Jochberger (06:25)

I mean, it's really different to what kind of executive you speak. Obviously, and as I come from manufacturing, I can speak best about this industry. So when you're talking to a CTO or a COO who runs the production and the production factories all over the globe or on various geographical distributions, for them, value is always attached to their business processes, so to the core value chain.

We are not talking so much about IT KPIs, but we are talking about throughput speed. We are talking about reducing downtimes. Now in the field of AI, are talking about reducing time to market. So how can AI help reduce the development cycles in R&D, which is especially a big topic for the heavy industries, also in the defense sector, where you have these huge complex systems that take years to be developed and then years to be produced. Meanwhile, this is really accelerating with the help of AI.

And of course, you can click, let me put it like that, when you have larger decision domains or an R&D domain where usually a P&L lies and where you have to take critical decisions day over day, then you have this beautiful underlying toolbox of use cases where you can click in nicely KPIs and the value to be measured.

And this is what we are currently discussing with a lot of clients. So mapping out entire landscape of use cases, ultimately leading into decision domains that then help take informed decisions for the C-suite.

Andrew Donaher (08:02)

Yeah, we've actually stopped talking with our clients about speed. The metrics haven't changed, right? Whether you're looking at your COGS or you're looking at your SG&A or your revenue per FTE or your insert KPI here, right, return on ad spend, whatever it is, your ROAS. We've gotten to the point now where we've gotten through that hype cycle and everybody's super excited about doing things quickly and we're focused back on those metrics. And we're seeing those metrics change faster and that's the critical thing. So if you're having, if you're able to build and test and deploy something in two months, you're able to realize the revenue change on that or the cost saving change on that in-year as opposed to a two year project. And that's what we're talking about in terms of changing the unit economics and the supply of the organization and how they're able to meet their demand for their customers differently. It's almost gone full circle where we went through this thing where we were worried about how fast things we get done, we're simply now worried about the core metrics once again and pulling the levers on those.

Geographic and industry perspectives of AI value

Helena Jochberger (09:21)

I think that's very interesting and I try to build on that what you're saying, because I think there's also a geographical component to things. I compare, you're sitting in Canada, so North American continent, right? I'm sitting in Europe and I think what we really see is that there are some geographical differences also. So I would say the mental model is somehow a little bit different. Whereas on North American continent, see more ship fast, capture value, fix issues along the way, can do mentality. What we see from the European clients is more like, you know, this hesitancy a little bit, prove it safe, prove it's compliant and justify before you scale. So I think that's also an interesting component to this discussion.

Christina Fung (10:09)

And Andy, what do you think? Like I think Helena mentioned a very interesting dimension around different geographies, different industries. And what do you see in North America, health care and government?

Andrew Donaher (10:24)

I was just thinking about that because having had, having been previously, as you said, on the national team and the global team, was able to see more of that. And now when I think about Vancouver and I, and I look back and consider how we're different in BC and even with some of the support I do across Western Canada into the Western US with different clients in different areas. I think that I used to say, I used to say you could never underestimate a human's hatred of change. And I've actually evolved that a bit and I said, and now I say you can never underestimate a human's fear of change.

And what we're seeing now is that people are embracing it. I think we have maybe a bit of an advantage because geographically, we're straight up from Seattle and we're straight up from the Valley. And so you've got this sort of community and this closeness to a community that is pushing the boundaries. And because we have, I would say, a very, very vibrant startup community here in Western Canada, what we're seeing is that sort of bleeding over into the healthcare space. It's bleeding over into the government space where we've gotten to the point now where we need to start telling more stories about the successes, about what's working, about what's happening.

There was one European country whose, I can't remember the name right now, Helena, you'll correct me, but one of the things that they had done was they had put QR codes on bus stops and QR codes on different government buildings. And citizens could scan that and they could then learn about it and they could do education courses.

And I think that's one of the things that we have to do more now of because I think that's what I'm seeing here is you've now got this inquisitiveness by citizens and customers on the good stories.

And I think we have to tell more of them because there's a lot of great ones out there. And the more that we do that and the more we help people understand, we'll get past that fear of change, no longer the hatred of change, but the fear of change.

There's some great work in the field of organizational psychology right now. If you're reading the American Journal of Psychology, there's some great articles in there and there's a gentleman that writes on that is Brett McFarlane. And he writes about how your intellectual, not your intellectual capacity, but your, stress load that you have, how it complements that. So the less stress that you have, the more you're able to accept that change and the more your community can do that. And I think right now that's what we're seeing here in the government and healthcare especially is people are getting over that fear and they're starting to adopt and accelerate.

We have one government client, I'll finish with one quick story, I know I've been talking for a long time in a row, but we have one government client here in Western Canada who they run a modernization product for one of their citizen interaction platforms, that's what I'll call it.

And they'd been doing a modernization off an old mainframe and off of some old Java stuff that had been around for decades. And they couldn't unwind it. They couldn't understand what was in it. They had a team that was on it for about a year and a half. Still hadn't done anything. We brought in a team using AI. We are code complete in two months and in production in four, where before they weren't even code analyzed in a year and a half.

So now we've got multiple pods in parallel, code complete in two months, in production in four, just because of the additional testing we're able to do around edge cases. And now the citizens have access to all these new features and all these new capabilities and the government's becoming more efficient in delivering those services. So some great stories there.

AI value beyond the balance sheet: Organizational change

Christina Fung (14:30)

I think that's a great example around value and objectives. And sometimes it's not necessarily just around task efficiency and things. And I use an example that when I was talking to a senior EVP in the bank recently, and his objectives for AI value is around innovation and is around innovating new products that can bring additional revenue streams to the bank. So it's not necessarily around like saving money here and there, operational costs. It is much more a growth mindset in that case. So I think it's really a great example of different industries and different geographies. And sometimes AI value is not necessarily only on improving efficiency and productivity. Well, having said that, we do see a lot of times come out our clients saying that, hey, I'm seeing productivity improved and there's more automation happening in the company. But the CFO typically will say, wait a minute, I don't see any improvement in my balance sheet. I don't see P&L impact yet. And we talk about people, we talked about a lot of change management, like a lot of these prerequisites, dependencies, important criteria for success.

What are the other things that, like what are the things you think is missing in that equation that they couldn't get to what the CFO looking for, like P&L improvement, balance sheet impact? What do you think? And maybe Andy can go first.

Andrew Donaher (16:13)

Yeah, I think sometimes it takes time for the benefits to come out on the balance sheet.

You know, for every, it's a three to one ratio, right? For every dollar you're trying to recognize in revenue retention, it takes three times as long as it does for expenses, because you've got to drive it through the market, you've got to realize the value. The other thing to look at is that that particular individual may not be seeing it, for example, is mental health and employee improvement.

So one of the biggest things we're seeing is the removal of non-value add activities, right? So people are able to focus on value-add activities. They're able to be more competitive. And what you're also seeing is I had two employees come to me. We put in a new feature. We helped AI with some traditional things. And I've never had this before. I actually had employees come to me and thank me, and they said, thank you so much, I now have the time back, my stress is decreased.

So all of a sudden that employee not only removed those non-value add activities, but the stresses and the particular challenges that they were experiencing that were bleeding over into their regular life has changed. And you're not going see that in the balance sheet in two weeks, you're going see it the balance sheet in a year and two years, but we're expecting to see these magical balance sheet changes in two weeks, that doesn't happen. But as your employees, as your retention rate, retention rate annually. So as you change, you look at your employee churn rate and as you're starting to look at your revenue per FTE increase. And the other thing I would say to those people, if their balance sheet isn't changing, that might be a good thing, because it means they're not losing in the market. And their competitors are adopting the same things, so they're not losing in the market. And this is a requirement for your ability to compete as you move forward. I don't know Helena if you’re seeing the same things.

Helena Jochberger (18:14)

Yes, I think to the point of the stress reduction, think this goes also hand in hand, both or tackles the usability on the one hand side and the process redesign. Because often we forget if we implement an AI algorithm and we train them probably, the outcome is as we have trained it, right? So we have that, we extrapolate it, and usually the algorithm does what it should do. However, if we don't do a proper process redesign at the end of the day, still the adaptability of the people might not be there, hence the stress level is increasing.

So that is having a closer look into the process redesign is really something “oldie but goldie” which we should not forget when talking about AI and AI values.

And then I could not emphasize more the organizational change management. We tapped in already on various angles of our conversation. I mean, first and foremost, it's the mindset. Having that tech for good always in mind. And since you're near to the Valley, this tech for good is a standing term that I would like to emphasize. I think there are marvelous things we could concentrate on rather than just, you know, going down the rabbit hole of doomsday AI scenarios, which we all do not want.

Because being a deeply philosophical person, I can only say the world is how we want it to be. And we all can work towards a joint future that we desire to have. And implementing AI in a way that is benefiting the people and society is something that we should strive for.

Christina Fung (19:48)

And I think that's a very good two points that both of you mentioned. So what do you observe when you work with different companies and clients that when these focuses shift, the shift of focus on the value and there are organizational impacts, people thinking about their operating model, their org structure, what are you observing and when you are having conversation with the client?

Helena Jochberger (20:19)

I would say from my point of view, it's quite a heterogeneous picture, right? Because the transformation progress of companies is quite different, so while some of them are quite advanced and ahead where they are already thinking about how could agents work alongside humans and how are we shaping this new operating model properly for the sake of the organization's future.

That is more, I would say on the digital leaders angle, while some of the other clients, of course, and also due to the nature of their business, they are more in the, I would say foundation status, you know, to do their POCs and to do their implementations and to road test if this really works out. So I would say we see both angles. How about you, Andy?

Andrew Donaher (21:13)

Yeah, I think that I'm just thinking through some of my different clients right now and heterogeneous, got stuck on that word when you said it because it's absolutely correct. But the biggest thing that we're seeing is the change in an organization's mindset with the need to get to production. You talked about being littered with POCs. That's, you know, something that, let's be frank, there's a value realization gap in the industry as a whole, where we've got, we're not starting AI off in this perfectly clean slate. There's a history of value realization challenges around legacy code, around POCs being involved in and never actually getting to production. There's a bunch of legacy around that.

And so we have to understand that operating models that benefit that are really going to be, sorry, that don't challenge that are going to be a challenge.

Helena Jochberger (22:21)

And I think also from the contributing parties or stakeholders that we have in the game, you mentioned the legacy from a technology point of view, but also various players coming, you know, to work with the client, our competition working with the client. And this is also basically an ecosystem that needs orchestration for the sake of the bigger picture, I think. So this is something I'm thinking a lot about recently, you know, it's a common market: thre are partners, there are competitors, but at the end, I think there needs to be this orchestration for the sake of the future of the clients.

Navigating fear and embracing change in evolving roles and organizations

Christina Fung (23:01)

And one of my examples is I'm working on some Client Zero within CGI of applying agentic AI in our own operations. And it is really a great learning experience. We talked about previously, I Andy, you talked about people's fear and things like that. And when we really do these transformation and engage the employees, understanding their roles, how it will be changed. Actually, what I find is they are much more positive because now they feel they're empowered with knowledge and knowing and what is going on and so in that exercise, only I think beyond the experimentation, we are actually doing it. We know that things will change, Like roles will change and what we learn today and versus six months later, I'm sure will be different too. But it's important to put it into action beyond the POC. To do it and iterate it and make it continuously better.

I think what Helena earlier talked about mindset shift. It is a mindset shift. A lot of people think “I need to know exactly what is going to happen and we have to do it perfectly.” And I think we all learn to be a little bit more agile, the lowercase “a” agile and be more adventurous, be more understanding that success comes in trying and success comes in experiences and not being afraid of changes.

So Andy, I do have a question for you that you talked about earlier about people's fear of change. You know, like we see it in newspapers and things like that. I'm sure our listeners also have similar questions like “Hey, everybody's saying the AI is taking over all the jobs.”

From our experience working with a lot of clients, it is different. It is not like that. Things are much more complicated and complex. And do you want to share a little bit of the deep insider thoughts since you're right in the mix of it every day?

Andrew Donaher (25:20)

Yeah, think there's a lot of, there's two things going on. I think there's, let's call it some “AI washing” happening, where organizations are maybe, AI's getting the blame for a few things that maybe it shouldn't be number one. Number two is, actually I just wrote an article on this, on the labor lump fallacy. So when you look at the labor lump fallacy in economics, it talks about people's fear of there just being a certain amount of work, and then once that work is done, there's no more work to do.

And I think it's really important that when we're talking to people, we illustrate that that is not the case. Like we talked about, there is not a finite amount of work out there. What we're seeing is organizations are now required to do even more than they've ever had to. And so that's the most important thing, is that if you're experiencing challenges around, or fears around those things, you have to look around at how much work there is to do. There's not many organizations that I've seen, let's say for example there's an organization with an IT budget of X. What we're seeing is that, that IT budget of X is actually not decreasing. There's just more work to do with it. That's why we always have project rationalization calls and project prioritization calls and program realignment calls and insert the next call name here that I've forgotten right now.

So we're doing a modernization project with a different client right now who has had the same challenge where they've tried this project four different times in history. They've had failures multiple times, never been able to do it. And now we're able to attack that like they've never been able to before. And we're able to actually now deliver that value. And so opening up that opportunity and opening up the ability for them to realize that value and that project that was never there is showing people that there's not the fear that they needed to have of AI taking their job. It's actually unlocking more opportunity. It's unlocking more work. And organizations that are acting upon that are the ones that are gaining their competitive advantages.

Christina Fung (27:37)

That's amazing. Helena, what do you see and what is your perspective about the fear of AI?

Helena Jochberger (27:47)

I also think we need to decompose a little bit the entire topic and break it down into little sub-topics. So let's start maybe by the jobs in that, in that sense, equal tasks, right? So of course, AI targets specific tasks, not entire roles. So what I see on the ground is that most of the jobs do not disappear, but they change in shape and they change in shape often quite quickly. Therefore this mindset attitude, this change ability is so important. What we see is a kind of enhanced polarization within the roles, I would say. So the routine parts of course get automated while still the judgment, the creativity and also the stakeholder-facing work expands on even the so-called edge cases. I've recently also written a blog article about AI values and especially when you're in these edge cases, think about the space industry when you're operating in space or near earth orbit or think about a pilot, right? Even if a lot gets automated, these edge cases where it's, literally when you're transporting passengers about life or death, these kind of roles stay human.

And two important aspects maybe, I think to Andy's point with the “AI-washing”, what we also see in the market is a kind of a new middle layer pressure. So middle management and coordination heavy roles are really being squeezed in reality because AI reduces the need for information aggregation and also for status tracking in its various forms.

And what I see, and I think this is also from a lens of a former global industry leader, quite important is this emerging of hybrid roles, right? So really what the market is looking for are people, kind of hybrid people who can combine domain expertise, industry expertise in combination with the AI fluency. And this is increasingly expected. So I would say it's a combination of many aspects.

But let's talk about the positive things.

Christina Fung (30:04)

One more possible thing I can share is when I was working with a group providing help desk services, operations help desk, one thing I actually feel very inspired by the people I worked with was they told me saying that, Christina, with these agentic AI solutions we are building here, I actually now can focus on solving much more important questions, problems for my customers who are calling me. And because now I actually can, I feel I'm delivering more value to my customers when they call and ask for help. So I think the AI value really also not only at the enterprise level, really is transpiring into the individual level of how much they can enjoy their work and find more purpose and meaning in their work as well.

Andrew Donaher (31:01)

Christina, can I add one thing to that? What we're also finding is, and this actually ties up both of what you were saying, but we're making the experts more efficient. We're not making laypeople experts. Right? And so when you're thinking about people's fear of their jobs and that non-value-add work and like you said, to have more purpose.

It's making the experts more efficient. It's not making laypeople experts. If you put me in a Ferrari and you put Michael Schumacher in a Ferrari, they're very different experiences. And I think we just need to make sure that we all understand that as we move forward.

Helena Jochberger (31:49)

And I really like your point, Christina, about how AI can be used as a purpose creator. I just wanted to underline that because often in our disconnected technological world, it seems like people are lacking purpose or lacking their passion. And I think while taking away more the boring stuff or the repetitive stuff and clearing up that mind for, let's say, more creative tasks, philosophical tasks, purpose can be found. So I think we should not underestimate that.

Christina Fung (32:23)

That's great. So one last thing to wrap up some of these practical experiences we have seen. We talk about people, we talk about technology. We also know that AI is not really about just technology. It's just a small component, not small, it's important component, but there are lot more other things that need to go with it. What other things do you want to add to what we have already discussed? I know a lot of times we talk about cloud, talk about data readiness and things like that. So some may be more technology related, some can be more non-technology related. So any thoughts to wrap up the topic?

The biggest mistake in waiting for your data to be ready

Andrew Donaher (33:06)

Yeah. And you just hit, you hit one that I wanted to, I wanted to hit before we tie up here is, people that say that you have to get your data in order before you start doing AI are wrong. You actually have to turn the AI on the data. So one of the things that we're seeing, the most success with across organizations is leveraging AI on the data to tie it together, to understand it, to clean it, to govern it, to enable organizations to find where it is and how different relationships can be made through their data.

And then from that, organizations are able to make more informed decisions, are able to make decisions about the various projects or programs that they'd like to add more fuel to. And I think that that is the number one, maybe not the number one mistake. The number one mistake is not getting going, but the number two mistake is not getting going because you think your data needs to be perfect. I guarantee you it's terrible. I guarantee you it's as bad as everybody else's.

And by using AI to help improve that and understand it, we saw one organization where we did that and the AI was right about the data definitions 80 % of the time and the data stewards only had to approve the definitions that they never had previously for us to move forward. I think that if there's one thing I can tell people it's that, please do that.

Christina Fung (34:09)

I totally agree with you. I was in in meeting with one of the journalists not long ago and I said exactly the same thing is around bringing AI to data and it's not bringing all your data to AI because it's going to take you forever just to clean up everything and so I totally agree with you and Helena, what do you think? What are the other, can be data related, or other aspects that we haven't talked about that's important for this AI journey?

The importance of leadership and trust

Helena Jochberger (34:57)

One of my other passions, besides philosophy, is of course leadership. And so I wanted to emphasize a little bit or underline a call to action to all the senior leaders out there. I would say, first of all, the best leaders aren't pretending to have all the answers, right? We're sitting all in that experimental boat called AI.

But at least they are clear about what they know and what they don't and what's likely to change. So I think walking the talk and being also role models in that kind of agility mindset is crucial for success. So that would be my first one, clarity over certainty.

The second one is honesty and really the transparency on impact. Also to the earlier question, we talked about jobs and so forth. So, I think it's quite crucial to be upfront and honest about how roles and about how the skills and maybe also the structures and the processes may evolve to the extent we know it at the time. And even when the message is uncomfortable, right? So I think this is something I want to convey. And then also this demonstrating the augmentation aspect of AI, not just the automation because you know what we discussed earlier showing employees how AI can really help them to be more effective, but also go beyond their own borders and also to have trust in these abstract transformation narratives that we all discussed.

And then of course, last but not least, the consistency and consistency between the words - what we say and what we do - because if leaders say AI will augment work, but immediately then cut the roles, of course the trust erodes quickly. And this is what I just wanted to have is some kind of wrapping up words.

Christina Fung (36:53)

That's great. I love that you brought up trust and, and then do what do you think about like trust and if there's some recommendation for you to leaders how to build trust in this case because trust has multiple dimensions in the in the AI dictionary.

Andrew Donaher (37:18)

Yes, I’ve got to say yes to that. The key to trust is you earn trust through action. You earn trust through your actions and what you do every day. And as Helena just said, if you, people want leaders who are going to help them to improve, who are going to help them to succeed, who are going help them to drive through their purpose, and you earn trust through those actions and moving forward.

You don't earn trust by sitting back and hiding. The currency of the future is going to be, or the competitive advantage of the future is going to be courage. And if you have the courage to move forward, I like clarity over certainty. I'm going to take that one away from today with me, that transparency helps to earn that trust. And that's where I think we need to learn how to move together quickly.

The different stages of transformation and the value realization gap

Christina Fung (38:17)

I think that's a great way of talking about what is really needed and the foundation of all these transformations and getting to the value. Organizational change is significant and none of us should underestimate the importance of it. So now looking forward to the future and we talk about value and our understanding of the value, people in different stages of their transformation journey, do you also see that as things go on in the coming months and years to come, the perception or what we talk about value will also evolve and change and if so, what do you think it will look like?

Helena Jochberger (39:03)

Yeah, I think in my mind's eye, you know, I always love semantics and good old Aristotle. So I, in my mind's eye, I put everything in three boxes. So somehow the early stage, the experimentation and many of the market clients, organizations we see, we have come over that, which is more focused on, you know, activity, outputs, number of pilots, which kind of use cases, more the model, accuracy and technical performance kind of discussions, where I would say from a value perspective, this is more hypothetical or very localized. So with what percentage could you predict the outages in your shop floor? So success would equal, does it work?

But I think in most of the cases, we have moved on meanwhile to the mid-stage or the scaling where we say, okay, we're shifting to operational metrics or even move beyond, as we heard also Andy say, so move beyond efficiency gains, move beyond productivity improvements or adoption usage rates and whatnot. So I would say at that point, the value starts to get connected more to business processes. And there I would say success equals more it is being used and is it really improving performance?

And then I would say that the third box is more the advanced stage, the value realization. I think we are between two and three right now where we are focusing both on business and on financial outcomes. So really in manufacturing terms, CEO working with CFO and what you also mentioned, Christina, revenue growth, margin impact, customer experience, customer retention. So a little bit more decision speed and quality. How do you see the world there, Andy?

Andrew Donaher (40:55)

Yeah, I'm just thinking through some of the things you said and I wrote a little something recently on the value realization gap. And I think one of the things that we're be able to do with more clarity as we're moving forward, we're seeing it already, is that the ability to draw linkages in the unit economics.

Helena Jochberger (41:19)

Mm-hmm.

Andrew Donaher (41:20)

And our ability to have more transparency and clarity, the ability to run more tests, the ability to run more scenarios, the ability to explore more revenue opportunities because we're going to be able to understand what's happening in two to three months or four months, not two to three and four years. And so I think that we're going to start to close that value realization gap. We're to see less tech debt. There was an article written quite a while ago now that talks about it's the title was, “it's time to start chopping wood with a dull axe.”

And so, you know, that talks about the, if it was Abraham Lincoln who said that, you could, if I had eight hours to chop down a tree, I'd spend six sharpening the axe. Well, now we're going to spend a lot less than that because we used to, you know, buy XYZ platform or XYZ tool and we'd sweat it. We drive the value out of that and we'd spend four years learning every knob and dial.

And I think now you're going to be switching up different types of software in six months. There's going to be more integration. There's going to be more need to do things because you're going to see more value quickly. And you're going to be able to have more transparency in what's truly driving value. And so there's going to be a lot more of that.

Christina Fung (42:44)

I totally agree. And one thing I also think you will see as it matures, it will go beyond of some standard financial related or typical operational metrics. I'm looking forward for some organizations will get to a maturity state that they can also measure value, the AI value bring to them in about the agility of the organization, their resilience to changes, because change is going to happen much more frequently in the future. So for the real health core of a company, think around building resilience, building trust becomes very important.

Innovation is going to fuel all these things going to happen, all the great things going to happen. So I'm also foreseeing that the value will also be extending to resiliency, organizational core health, both building agility in the operating model, ability to pivot quickly and adapting to changes. I think all these will be great AI value to the company.

So last but not least, a couple of things. So you guys are all very creative and staying at the forefront with clients and in the AI world. What do you think the next six to 12 months are going to look like? Any crystal balls in front of you?

What’s next in the next 6-12 months

Andrew Donaher (44:16)

I don't know if we have enough time for that. I think we're at the point now where you're going to see an incredible amount of acceleration workforce modernization is I think what you're going to see. The adoption and the changing and the way we work. I think you're going to see an incredible amount of change right there.

If you think about it, AI right now is the least effective and least productive we will ever see it in the future. And so you think about how organizations are going to reorganize and how they're going to be looking at modernizing your port, how you’re looking at folding in quantum computing to the optimization of your logistics with AI. Those things are real right now. We're already seeing clients jumping on that. So I think how we work, we're kind of past that what is it stage and the adoption and the changes in how we work over the next six to 12 months are going to be significant.

Christina Fung (45:25)

Alana, what do you think?

Helena Jochberger (45:28)

So my core message would be don't start only with the tech, don't start with the technology, but start really with a decision you want to improve. You just touched upon resiliency, Christina. And I think that's a very big topic, especially when we talk about supply chain, the function of supply chain, or even think of a state, think of a society where you have the citizen supply chain, right?

So, I think the organizations or even societies that will be getting real value from AI are not asking where can we use it. But the question would rather be where are our most important and repeated decisions that we need to take and how do we make them better, faster, more scalable, more acceptable by the users. And I think there is a shift in these few things.

It ties AI directly, of course, to a business value, in that case, supply chain and not just experimentation. And it forces us also to get a clarity on what humans and what the AI should do. So this mixed, you know, AI agent working alongside human scenario. And I think this will eventually and naturally drive the adoption because it's embedded in the real work. It's embedded in a real decision domain.

And so for me, the practical thing would be pick one high impact decision in your business and redesign how it gets made with AI and then expand and build it from there.

Stay curious and brave, and act now

Christina Fung (47:03)

So last but not least, just leave one thing to our listeners today. It can be one recommendation, one advice, one lesson to learn to share. What will be that one thing? And why don't we start with Helena.

Helena Jochberger (47:19)

Stay curious and brave.

Christina Fung (47:23)

I love that. Andy, what’s the one thing?

Andrew Donaher (47:27)

Act now. That's it, act now. Move. There's opportunity. Your workforce is looking for it. Your customers are looking for it. Your citizens are looking for it. There's no shortage of things to do. Act now.

Christina Fung (47:44)

I totally agree and the one thing that I would advise would be try and stay positive, be the partner of, make AI be your partner and don't fear it, so summarizing some of what Andy and Helena said. And that's a lot of great insights and thank you so much Andy and Helena for sharing. I hope that not just me, but all the listeners here are taking these valuable thoughts, insights, experience back to their practices. And we look forward to hosting more of these podcasts with both of you, Andy and Helena. And thanks for the listeners for staying with us today. Thank you.