| If you don't have an account, Register here |
May 26, 2023
Moderator:
Phil Fersht, CEO and Chief Analyst, HFS
Panelists:
You can listen above or watch this HFS Videocast here:
Session Description:
In our plenary session, join this high-level debate among the big-shots of the business tech world to shine some light on critical topics up for debate, including:
You can also listen on:
• Apple Podcasts
• Spotify
• Google Podcasts
• Anchor
This transcript was auto-generated from the original recording and lightly edited for readability. We've done our best to catch errors, but names, technical terms, and company references may be misspelled or imperfectly captured. For the definitive version, please refer to the original audio or video. Views expressed are the speakers' own.
So without further ado, I’d like to ask our esteemed panelists to come to the stage. Let’s get into some serious conversation about which enterprise leaders will succeed and who’s going to fail as we tackle this global assault on everything we knew was stable. Maybe Manish, from Accenture, do you want to kick us off?
Yeah, I think this is the most often debated question, right? And the key thing I will say is that it’s the leaders who will embrace change, who will focus on changing the way the work gets done. The leaders who will focus on data as a primary thing — because you can’t have all the stuff that Phil talked about if you don’t have the basics on the data really sorted. But embracing change, changing the way we do work, data at the core, a digital core at the center of everything we do — that is the key.
CP, do you agree with that? How has the role of leadership changed, in your opinion? Is it going to keep changing?
You know, when I was growing up, I remember my father telling me: work hard like an ant, be strong like a tiger, and be friendly like an elephant — where a child can also go to an elephant and have the right attitude. And somewhere, if I go back to my father’s lessons, which he was trying to paint to a child through the animal kingdom, I think the fundamentals have never changed. So whether it is medium term or long term, on the role of leadership I would only add to what my father used to say: have the vision of an eagle. I would further add that you lead in a formation like geese, and you communicate the way dogs or wolves communicate. And why am I bringing the whole animal kingdom into today’s conversation? Because nature has always taught us a few good things, and my personal belief is that it’s the leaders who have recognized — whether it is Satya Nadella coming in and recognizing that the old method had to change — what Satya did was actually adapt himself. So, again, maybe not the best comparison, but being the grandfather of a two-year-old, I’m only reminding you what my father taught me. So I’m back to the animal kingdom.
Good stuff. Maybe Sander — you’re in the talent industry these days. What are you looking for in leaders, and how do you think that role is shifting?
Well, I think the role of leadership as such hasn’t changed much. I think how leadership is executed, if you will, has changed a lot. And we’re all about talent that runs up. If you just look at the world of talent, talent is scarce, and that is not going to go away — unemployment in all the major markets is at historical lows. So, as companies, as leaders of organizations or businesses, we have to make sure we create an environment where talent can thrive. That means you need to give talent, to a great extent, what they are looking for. I was on a panel the other day in the Netherlands — we used to have a saying, “for you, ten others.” In other words, if you don’t want to work for me, I can find ten other people. That’s no longer the case. So you almost have to start treating your employees as customers. You need to really look at what they are looking for. And what are people looking for these days? They’re not just looking for a job and a pay slip. Flexibility is something that, post-COVID, has risen to the top of the list, and that’s not going to go away. You need to create an environment where you can provide flexibility and the right work-life balance. Then, of course, it’s about meaningful work — is the purpose of the company aligned with what your employees are looking for? And last but not least, and at Randstad we call it equity — we want to be the most equitable talent company in the world. Meaning, do you create a work environment where people feel comfortable, where they don’t have to leave anything at the door, where they can focus on work rather than on other things playing a role in their lives? Really, you need to put yourself in the shoes of your employees and create the right environment, because otherwise you can be the best leader in the world with the greatest vision, but your people are not going to follow you.
Excellent, thanks. So, Tony — working in a large enterprise as a CPO, do you feel that enterprises are adapting to this new world, or are we still clinging to the rules of the past?
We’re definitely at the inflection point, and what we’re faced with today is very different from what we’ve looked at in the past. You pointed out cybersecurity a moment ago — it’s something that has always been prevalent, if anyone has read up on the Lazarus group or things like that. There are some really serious issues out there. But from a recessionary and post-pandemic world, we’re now looking at where we work, how we work, and we’re also looking at new employee groups — this is a different employee group today than it was 20 years ago. The old world of how we used to lead isn’t the same any longer. Are they adapting, and how are they adapting to it? I think they’re asking managers to learn how to lead more with their heart than they’ve ever had to in the past. And in terms of the rules of the past — being locked in our offices from 8 to 5 is a different piece of it. But when you combine that with cybersecurity rules, it’s almost like the BPO age of 20 years ago, when you’d lock all your work in a room and no one could bring paper in or out and you disabled hard drives. It’s a very different world, and we’re going to be facing that everywhere.
Excellent. And maybe, Suzanne, from your perspective — you run the Americas business at Wipro. Do you feel we’re at an inflection point, or are we clinging to the past too much?
I absolutely believe the comments around leadership are critical, because you’re only as good as the team that’s behind you. And as we think about the type of talent we need, we need to help them move forward in terms of the training we provide and give them a sense of purpose. I truly believe in this purpose-driven culture — having someone just generate a piece of code, a bot can now do it. Those are the types of skills that are most at risk and can most be done by AI. So we need to really help our people adapt to those new technologies and have a purpose for what they’re doing. And going back to the point on flexibility — that’s a huge area we need to make sure we adapt to as a business, to give that flexibility to our employees.
Thank you, Suzanne. To build on that — Sander, I’d love to hear more from you, in your role running one of the world’s largest talent organizations. What steps can businesses take to attract and retain the best? And maybe a little about the flexibility of working remotely — how do we make flexibility equitable?
Yeah, so at Randstad we talk about flexibility with intentionality. Meaning: we are here to do a job, that’s number one. We’re here to serve our clients, we’re here to serve our talents. So if the work requires you to be in the office, or in some physical place where you collaborate, then that is what you have to do — you cannot say, sorry, I have to walk the dog. And outside of that, we let the teams decide where and when they work, how they work, because that is where the flexibility comes in. I’ve seen many of our clients struggling to say, well, we want to be four days, we want to be two days, or even five days — and at the end of the day, with those types of measures, everybody is frustrated. Leadership is frustrated because people don’t show up; they simply don’t. And employees are frustrated because leadership is telling them what to do. So finding that balance, which I think is different for every company and even role by role, team by team, is really important.
OK. Some data we were sharing from a big study we ran recently showed — when we looked at what employers wanted to do versus their employees, you can see the employees on the right — the biggest difference was people wanting more empowerment to make decisions. And people were less focused on the missions and values of the company.
Yeah, and something interesting to add here: people are increasingly willing to decide to leave the company if what they are getting is not what they’re looking for — whether it’s around equity, flexibility, or purpose and values. Fifty percent of the people in our research say they’d be willing to leave the company even without having another job yet, because people have options. We’re talking about the recessionary environment — I think we’re sort of on the brink of recession, but unemployment here in the US is still at 3.5 percent. When we all went to school, at least when I went to school, 5 percent was the perfect level of unemployment. So it’s a different world out there.
Certainly is. Are we going to recession? Hands up?
Well, there was a hurricane coming — Jamie Dimon said that three quarters ago. I haven’t seen it yet. So my best-case scenario is that we’re going to be sort of fooling around a little bit; that could be two, three, four quarters, I guess.
Yeah, a global shortage of talent may be staving it off.
I want to make a comment. Most of us — including the charts you presented — talk about the global shortage of talent. I start with the hypothesis that if there are seven billion-plus people, and in most countries you visit unemployment is one of the big issues, then why is talent shortage an issue? You put in one of your charts that you need prompt engineers and you don’t have prompt engineers. Now, if you anticipate the need for a prompt engineer — and most of us have had to set up our own universities; Tech Mahindra has its own university, every company has its own university — if you anticipate, and if you go where the people are, if you take into account the new technologies, remote working, online and offline learning, there is no talent shortage. It is just that we are not decisive. We do not anticipate, and we do not know. How many of your staff have already said, “I want to become a prompt engineer”?
Sorry, what was the question?
How many of your staff want to become a prompt engineer? As I said, it’s my job to educate people about what the trend is. And, trust me, I’ve gone back not only on prompt engineering — I’ve actually taken them back 30 years and said, start with the CASE tools. Because you can’t become a prompt engineer if you don’t understand the basics and you don’t understand the CASE tools.
Absolutely true. So let’s think about this — we’ve been talking about the current business environment, and how is this impacting tech and innovation, what’s getting back-burnered and what moves to the forefront? Chetan Dube, the founder and CEO of Amelia — a very well-known intuitive digital employee — I’ve had the pleasure of observing the development of that capability. What is happening now in terms of the speed of the shift with tech and innovation, and what do you think is going to fade versus what is coming into play?
Well, I think we are in the midst of what we’ve been talking about — the digital Darwinistic curve. I remember 2015, Sander being in my offices, and me asking him: what if digital employees were to become indistinguishable from human employees? And I asked Sander, what if it works as advertised? And Sander said, nothing works as advertised. But the time has come. We said that by 2025, digital employees will become indistinguishable from human employees, and proof positive of that fact, the time is here. Now the question is, how do we face this trend? Will incrementalism cut it? If you subscribe to Phil’s philosophy of the S-curves, incrementalism in a period of digital Darwinism, where the inflection curve is much steeper than the curve you’re planning on growing, is not going to cut it. And so generative technologies, as Phil pointed out, are an absolute quantum shift in the way you operate. And, by the way, generative technologies are not that “oh my God, it just happened right now.” Everybody that’s been following it — and I’ve been doing this research for a couple of decades — you’d go back to the ELMo models, then to the bidirectional BERT models. What were they? They were just a few hundred-million-parameter models. What has happened is that the 345-million-parameter model graduated into Turing-NLG and Microsoft models of about 10 to 17 billion, and that graduated into GPT-3.5, a 175-billion-parameter model, and then over a trillion-parameter model. The hallucinations and all these effects will diminish. You’re going to find them to be indistinguishable from humans. The challenge is that what they’re providing you is the equivalent of what you and I have talked about — a whisper agent. They’re providing you assistance; they’re providing you a recipe for the cake. Can you have the cognitive capability to bake the cake? Can you raise generative to the power of cognitive? Can you take it from a system to resolution? That’s where the next big wave will be, and that’s where you’ll see the forefront of research. And what’s back-burnered? The metaverse and even some of the ESG initiatives have gotten back-burnered because of the GPT wave. But sorry to bore you with the numbers and math — I think it is very compelling: as you see the exponential rise of the parameters in these models, you start to see the hallucinations drop, and you start to see them asymptotically approaching human levels of competence, and that’s what you’ll see more pronounced in the next 18 months.
Can I ask a question? Chetan, in Mumbai around 2015, 2016, you made a statement that 70 percent of the current IT jobs would get automated, and it became an Economic Times headline in India.
That’s right.
What is your forecast now? Because between 2017 and 2023 the market actually expanded, so the jobs did not get eliminated. What’s your forecast for the next five years?
Well, let me ask you — sorry, I’m doing your job. First, I’ll say: Tech Mahindra, I submit that 70 to 80 percent automation is almost a floor of what you do. So it has happened. Almost all incoming exceptions — if you’re a competent enterprise, of the total incoming volume, 70 percent is now being automated. That is being done. Now, yes, you’re right again, CP, that the volume has expanded. More and more has moved from ITSM to ESM, everything has expanded and everything is becoming IT-centric, as this curve also showed. The budgets are expanding. But this curve is shifting from just automation — which was often a back-office premise — into being front office, being customer care, and being equivalent to human beings. So it’s a much more pronounced change, as opposed to a back-office kind of thing that people in the audience have perfected at this point. So how would you blueprint your autonomous, your generative enterprise? Would you build it with humans? How would you give the roles to humans, and how would you give the roles to the generative digital employees you’ll have in your enterprise? The leaders have to face that challenge of what the roles are, and the hiring has to focus on the quality of employees you get, as opposed to the quantity. That’s a big shift in the current philosophy.
No, I mean, you know what’s going through my mind? When I sit in a car, I know maps have got automated, but we still — some parts, yeah, we have done the automation. And you have driverless driving; maybe generative AI is bringing us more into driverless driving. I’m sorry, I’m just using my visualizing. But remember I told you the animal kingdom. Time and time again, we’ve had the discussion that technology is going to replace humans.
But history has also taught us there will be new jobs and new roles to be played, and the demands of us as a society, as consumers, as businesses, are going up every time. So I’m not concerned about there not being enough jobs. I’m more concerned — and you, CP — about how we’re going to ensure that the people who have to do the jobs are going to have the right skills. That, I think, is the critical question. The first time we were told that jobs would disappear was not in 2015; it was 1851, 1871, right? I was only reminding NASSCOM — Chetan did a keynote at NASSCOM in Mumbai, and I was reminding Chetan about that keynote. That was just to scare you all a little bit.
The point I keep remembering is that we just went through COVID, where every business decided it needed to be digital. So we saw this massive boom, but now we’re recognizing that the skills are going to need to change, because people — to write code, check code, contact center, frequently asked questions — all this can be done by a bot, through AI, and the AI is learning at the enterprises they support. Whether we call them people or workers, how do we have the talent that can teach it in the right way, remove the bias, and make sure it’s actually adding value to our clients and to the enterprises we support? That’s what I think needs to be the focus as we shift our focus to talent.
Actually, I think the biggest issue is that our labels of talent still refer to 10 years ago. We are looking for jobs that are composites of tasks similar to what they were 10 years ago, with all of the technology that’s now available. Those jobs need to be redefined, because there are tasks in the jobs that are not needed — which then links to talent retention, because nobody’s stopping by and asking people what makes them happy about the work they do. We’re trying to fit them into a box. And that’s what the whole concept of flexibility is about. So if we can re-craft the jobs and build them with all the available technology inserted, all the questions about how we retrain people and redefine skills become much easier to answer. The last big spike in spending we saw in IT was all tied to pandemic time, implementing all the technology we needed to automate the work. Now, what we’re talking about is doing the work. It’s not moving the work from widget to widget; it’s doing the work. We can now capture every piece of data in an email and combine it with all third-party data. We’ve never really spent a lot of time focusing on how sales sells, how procurement procures, how marketing markets — and now the technology we’re talking about is that type of technology. It’s not the work of moving widgets around the globe. Because of that, what’s getting back-burnered right now is all that workflow. It’s all done, it’s all advertised, we’re carrying it across in the depreciation of all our IT budgets. We don’t need more of that tech. We’re really asking the question now: we’ve got all this, what do we do with it? What do we do with what we’ve got?
The metaverse was the flavor of last year, and now it’s all about generative AI. Are we on a constant hype cycle, or do we expect these innovative technologies to reach their promised heights? Maybe Manish, what do you think?
Hype is all about hope — you should always remember that. Why is hype created? Because we are all seeking higher profits, higher productivity, higher almost everything. I think some of the earlier technologies didn’t have good underlying technology to create the right adoption. We were on automation — I see Mihir here sitting in his office; I don’t know whether it was 10 years back we were talking about RPA and everything else. Having said that, the generative AI we are seeing, I think, is real. All the other stuff we’d talked about never had the underlying technologies. And even for the gen AI stuff, we’re seeing it already — I don’t know how many of you saw the slide that was presented, the 30 percent and now the compressed transformation around 30 to 70 percent more. I think that’s real. This is probably more real than anything else I’ve seen, just because of the underlying technology now. And when we talk about this, the one biggest hindrance to greater adoption, which people might not look at seriously but I think will come, is the quality of data. We can create all the stuff, but if the quality of the underlying data isn’t there, we will struggle even with gen AI. But organizations are realizing it; they’re building the digital core. And look at just the adoption we’re seeing — if you’re creating your own private version, I think it is real, and not anything like the previous ones.
Yeah, I would agree, it is real. Just building on what Chetan said earlier, step back and think about what’s possible. What’s possible here is to deliver differentiated services personalized at the endpoint, which didn’t exist before — whether it’s healthcare or education — and then it is about scale. This is not a lab experiment that’s going to happen in a corner; it’s accessible to all of us, so it drives scalability. It changes the equation from one-to-many to a many-to-many equation, which wasn’t there before. And I agree with Manish: if the underlying data is not sorted, this will not scale. So if there is an emergency measure required right now, it is about the underlying data — recognizing what data is needed, what decisions leaders need to make, what data is required for those decisions, and how you drive automation and better processes sitting underneath that data.
It’s interesting, because I’m getting the sales pitches all the time from different suppliers about what works and what doesn’t. The comment around data quality is so real, because if it’s only 80 percent accurate, for those of you who run operations, you do not want to put a bad transaction through your systems — you’re going to be wrong 20 percent of the time. This stuff is real because now it’s presenting not just 10 or 20 percent of the bits of data, it’s representing 80, 90 percent of it. It’s the value of the human in the loop that we haven’t spoken about yet — that’s really the point. When we look at underwriting in our world, how can we take all this data and improve underwriting decisions? If you’re missing key bits of data, it can be crucial to making a bad decision on a risk. So it’s about how we resist the lessons of “oops, we made a mistake,” and put processes in to manually review the data to get back to 99.9 percent quality. We need to find that balance, and as this data gets better, I think we’ll get a little bit better as humans in the loop, being OK with errors. When I think about the metaverse last year, there was close to two billion dollars invested in it, and clients were still trying to figure out the right use case — learning and development, digital twinning, cloning, conferences — things that seemed useful, but almost looking for that next use case to drive real value. I think it’s the combinatorial things, the metaverse and other things, that will reach their value, and it may take some time — just like when we had cell phones, they used to do nothing but phone calls, but look what we can do now. But when you look at generative AI today, it’s already disruptive. I saw it from my kids before we were even talking about it — they were using it to write papers, and they were the human checking in and validating it. It’s already disrupted schools, and now we’re seeing it in the workplace, and the use cases are abundant and game-changing. So I say it’s already disrupting, and it’s about how we make it valuable and useful, and making sure the data quality is good, because that will be the key to everything.
One point, Phil, before we leave this — we talked about the data, which is great, but I think responsible AI and the compliance towards this is going to be massive. The first thing we did as an organization was around compliance and what is responsible gen AI, because if you’re going to create biases, wrongness, and inaccuracies, you’re going to create havoc. So you need to put those parameters right before you start going too bullish on this.
Yeah, little things like being able to cite the sources of data are huge — where is this stuff coming from? We had a meeting in our research teams thinking, oh my God, a junior analyst could just pull some junk from anywhere and we’d think it’s great work. I just look at the developments going into GPT-4, and there’s a significant shift — it’s almost like an iPhone moment. When Steve Jobs released the iPhone in 2007 — and I suggest you go on YouTube and watch his launch — he had the next 15 years mapped out on that phone; he knew how we were going to adopt it and the speed of adoption. I think we’re at that point now, where we can see this happening and see the potential, and we can see what’s failed. Now, I don’t think the metaverse failed — I just don’t think anyone wanted to wear a bloody headset. If you’ve walked around bumping into every item in your living room, you’re like, OK, that was fun, but I’m not doing that again. We are at an inflection point. So let’s think about this: does this create more opportunities than threats for enterprises? Suzanne, do you want to elaborate a bit on what you said — is this a real threat or an opportunity? I noticed in India, for example, I see less buzz about this than anywhere else, because some think, oh my God, you’re going to start programming using this stuff, we’re in trouble.
I look at everything as an opportunity. You can say it’s a threat — a threat to what? So the question is, how do we change how we work? And I really do think it goes back to talent and people. The reality is that standard processes and repetitive work were all going down the path of being disrupted, and that will continue, so people need to rethink the type of skills they have and the roles they’re going to play. You’re not just going to have something generate code and throw it out the door; you’d have a human, and there need to be controls and governance around quality. And there’s the human element that we need to make sure of — that’s responsible, whether it’s AI doing code checking or code development. That’s an important part of how this moves forward: the responsibility of organizations. We’re already putting out responsible-use frameworks in the organization to help clients think through how this gets applied and how you use the right learning models and make sure it’s adding value. So I think it really goes back to talent and how we rethink how we work and the jobs we do, because, candidly, I think it creates a much more exciting opportunity for people than the kind of work some people are doing that can be quite laborious and boring. So to me it’s an opportunity, but it’s going to be a challenging one for leaders to navigate.
I think it creates an opportunity to link the work we do with real outcomes, and it brings outcomes closer to the work we do, because it connects data, process, people, and technology with the outcome the enterprise is trying to achieve, which is your client. So to me that is a huge opportunity, which was distant before generative AI.
So, what Chetan mentioned again — the doubling of the number of tokens from 3.5 to 4, and the number of parameters from 750 million to more than a trillion — if you look at those two data points, Phil, I’m sure when you do this in two years’ time we’ll be talking about GPT-10, and we’ll think, not sure what GPT-4 was all about and why the hype. It will learn and it will get better. I’ll repeat here again: the issue will be that if you’re going to use it in a reckless way without any responsibility, you’re going to land large enterprises in trouble. So I want to remind everyone again of the question around compliance and responsible AI, because if you overcome that, then you’ll probably be a winner.
I mean, Sander, how should leaders prepare their workforces for this?
Well, if I’m just looking at this chart, it says “the generative enterprise, AI-driven.” I fundamentally disagree with that, because it will not be AI-driven. It will be people-, talent-, skill-driven, because ultimately it’s about what we do with the technology. And there are maybe two big concerns around talent. One: we all grew up learning basic skills — writing a paper, doing market research, whatever field you were in. Now, if AI does all that, how are we going to train our young folks to learn those basic skills so they can ultimately be the judge of something that is an AI output? This is the one big thing going through my mind, and I don’t have the answer. The other thing is, if you ask people who is responsible for their career, more than 50 percent say it’s their employer. That’s a big issue, because people need to take charge of their own careers. And for us as leaders in businesses, it’s not just a matter of putting a catalog of online learning somewhere — we’ve done that, and we found out people go online and see 350 courses on marketing and ask, where do I start? People need help. They need to be told: you’re here now; if you take these two or three courses, you can aspire to that role. They need to see those pathways, which is really important. And these pathways are obviously changing in a world where AI is so important. But I don’t think things will be AI-driven; I think things will be leader-, people-, and talent-driven.
Yeah, nothing’s more powerful than when people introduce something to the enterprise. Everyone wanted to buy Apple Macs and brought them in, and eventually Microsoft gave up and said, OK, you can run our software on them. It’s the same with this — if our kids are using ChatGPT and using it to run their lives, they’re going to want to work with it, and it’s going to come in that way.
Exactly. Just one comment on that — I agree with what Manish and Suzanne pointed out, that the leaders have to think about what the new enterprise would look like. I think we have to assign the roles of humans, the human assets, to non-commoditized, creative parts of work. Take investment banking. You may want to do a classic evaluation analysis: I want to find the P&L and balance sheet, read the cash flow, compute discounted cash flow, and compute the comparable company analysis. That’s all I want to do. You go to GPT-4 and type that in, and you’ll see a narrative. Now imagine if you had an actual language compiler that could read that narrative and had a cognitive system on the back end that would actually do the discounted cash flow for you and compute it. You don’t want to be a Goldman Sachs or a JPMorgan and say, I’m going to hire a horde of investment-banking associates to do this classic work; you want to reapportion them into non-commoditized areas where they can be an asset. The role of humans has to be rethought in this new world, when you raise generative to the power of cognitive. You will be printing out digital employees, and we have to rethink the role of human employees so they can be an asset to make our organizations go further. It’s the responsibility of the leader to do that, because this will come very quickly in the next 18 months. And CP and Sander will hold me accountable to it, because Phil has a tendency to bring the big giants and put me on the stage, and they beat up on me. But we all love that. In the next 18 months, if I have the privilege of sitting next to these giants, they’ll hold me accountable to the fact — are we really looking at digital employees doing investment banking? And you’ll say, yeah, actually they are.
I was just going to say — you said digital employees. Does that make the other employees more empowered? Because we just learned that empowered employees are what retains people; people get excited about it. Does this enable people to not just feel more empowered? I say it does, and I think that could be the real game-changer for us as we think about how we retain the best employees.
Yeah, but the question is, you put a machine gun in their hands and what are we going to do with it? That’s a real risk. Bad stuff can happen, let me put it like that. For the last 20 years it’s been the global redistribution of the workforce to different countries for low-cost labor, or being able to use ERP systems to do faster data and reporting than ever before. Now we’re talking about what people are going to do, doing the work and how they’re going to do it differently. How you learn the work is going to change — it’s not going to be through online learning, that won’t be fast enough. That’s going to be the big challenge: how do you learn how to do a job when none of the beginning jobs are available any longer? How do you step into a knowledge-based learning role when there’s nowhere to learn? How do you start?
But I actually challenge your statement a little bit, and I go back to that 1851 comment Manish made. At that point in time, if you were an artisan, you were really valuable, because there was no machine available to do anything you could do. You defined the work as being an artisan and then asked, oh, what is this person going to do? But look at what has happened. So the outcome of a job itself is going to change. But I get your point. I can now use a 3D printer and make my art, but someone still has to design the art.
Not necessarily. If I want to send you a piece of art and I can read what you really value, I can create multiple pieces of art in a minute, and you’ll say, hey, I like this and I don’t like that. So that didn’t exist. You’re 100 percent right in that way, because now we’re looking at these computers actually making artwork, videos, movies — right now we’re talking about output being purely data and transactions, but what if it was “make me a movie, please,” and it actually shot the movie for you live? The Met recently ran a challenge to critically appraise every piece of art in the museum, and what it came back with was that it was able to identify pieces that human eyes could not, and links between various forms of art and common themes — this wasn’t even an ask. So just think about the changes to the outcome, and that’s what’s going to happen.
OK. CP, you had a point.
From my perspective, the biggest challenge is not the skill development of employees. The bigger challenge is how leaders will orient themselves to the tectonic shifts happening in the market. Phil, you started by reminding us of the Apple Mac in the enterprise. How many of us remember when the tectonic shift happened as we switched from rotary phones to mobile phones? It just happened — we didn’t have to go through anything. So my personal assumption is that for employees or future employees, because these new technologies are so adaptive, they will learn on their own. It’s the same way a one-year-old child was never taught how to use a smartphone — they just learned it. I think the biggest challenge is: will leadership be able to adopt, and will they be able to give direction to the new world? Will a technology company realize that domain knowledge will become a lot more important? Will a technology company realize we need to work more on the use cases now, instead of just saying I’m a tech shop? So that’s really where some of us as leaders will have to question ourselves.
Completely. It depends. The BlackBerry completely transformed things — remember, it came into the enterprise and every leader had one; it was your privilege, and you spent all day on it playing that brick-breaker game. It completely transformed how we interacted with each other, but then better tech came along. This feels very different, because the tech’s coming along and it’s figuring out how to change your mindset and use this more effectively, and then you start to try it out. I’ve seen people using these voice apps with chat already, and I don’t even think we’re going to be using keyboards much in two or three years’ time. I think we’re going to be talking to our machines and using a lot more voice.
OK, so in the interest of time, I want to have a quick conversation about partnerships, because there’s a big services relationship theme in today’s conversation. Maybe, Tony, you want to start us off — can we get past this old world of effort-based partnerships? One slide says we all want to buy on performance, but we still buy on FTE. Everyone says they want performance, but then they buy FTE. Is that procurement’s fault, or is it — I don’t know whose fault it is?
I think the economic value being presented right here, and the value being asked for by the suppliers, seems to be — the suppliers seem to still be asking for outsized amounts compared to what the buyers want to do. There are so many innovative solutions available in the FinTech and InsurTech areas that I don’t necessarily need to get it from a big-name company any longer. There are lots of choices for us as buyers to get it from different places where I don’t need to give up millions of dollars in a transaction. I know Accenture looks like the big 100-million-dollar diamond clients, shooting elephants out there, but there are a lot of solutions that don’t require big implementations nowadays to get good results. So I do think that’s the challenge. Will we ever get down there? It’s whether the companies themselves want to, and how they see their internal incentives aligned to the decisions they’re making. If my goal is to take out cost, I’m going to take out cost. If my goal is to make revenue, I’m going to combine and partner for revenue. It’s just those types of definitions.
This has been discussed for probably a decade, right? Shifting from effort-based to performance-based has been a challenge — whether it’s the procurement organization trying to get to the lowest cost rate. I do think that in the last few years, in fact since COVID, there is greater adoption of what in Accenture we now call total enterprise reinvention. We’re seeing a massive uptake where people are not just doing finance or IT — they’re trying to take the full thing, what I call creating functionless organizations. So there is a trickle toward the performance measures. On the adoption — people say they’re doing all these big changes. We did a research study: 86 percent of companies said they are doing big transformations. We dug into it, and only 6 percent were actually doing what we call total enterprise reinvention, touching every part of the organization. Siloed finance, HR, some IT — but nobody is taking it holistically, and that’s why I think that once you start adopting this, the curve will change.
Toward more performance, yeah — and also the definition of value. Is the value cost, is the value revenue, is the value new customer acquisition? We’ve now got contracts where the value is the full end-to-end — make sure the product is sitting on the retailer’s shelf on time and in full, and if it’s not, we don’t care which function it comes from, you run it, you are accountable. So one is the definition of value. Two is the level of maturity of the buyer — is it a functional buyer, is it an enterprise buyer, and what are you looking for?
Yeah, exactly. It’s like having a salesperson and incentivizing them on a personal target versus the whole company target — getting them thinking much bigger and broader: how do I wake up in the morning and think about how I can help my company?
Just one comment I’d add, because technology is now the same as business, this becomes harder and harder. If it’s about my finance department — I have 100 people, I want to do it with 50 — that’s easy. But technology in business, growing the business, new services — you need to be agile, and it’s hard to pull this off between two companies. It depends. So if I’m going to sign up with Randstad, I want to make sure I have a great influx of talent to my company — you’re going to think about the quality of talent. If I sign up with Weber PR, I’m going to think about, am I getting quoted in the right publication, the Wall Street Journal, that sort of stuff. It’s very simple: you’re looking at outcomes. And if I’m signing up with whatever to get IT support, aren’t you thinking about the great IT support you’re getting, or are you thinking about how much am I paying for this? It’s as simple as that.
So let’s get on to the final question, which I really want to talk about. If there’s one wish you could make to change this industry for the better — just one wish — what would it be? Maybe from everything you’ve thought about today. So maybe we’ll start with Tony and move down.
I was going to start with “no PowerPoint,” but…
That’s a good one.
Yeah, I wish it weren’t a part of it. But if there’s one thing I could change in the industry today, it would be that I wish the sales hype would go away. Hearing all this — it can do 99 percent, 100 percent — we need to be very practical. So I wish the hype would die down a little bit and we could be very practical about what the real effectiveness is today, and how we as buyers can adopt it and learn how to use it over the next few years and start getting better at it.
So, Phil, you talked about the generative enterprise. I’m going to call it a generative ecosystem, because there is no single enterprise that’s going to be fully capable of delivering everything that is needed. But if you get a generative ecosystem, you will automatically look at outcomes.
There you go. I can see it’s pricked up over there.
I think complexity. I’d love to see complexity go down and value go up across the board — across everything we do in our ecosystem — because I think it would just make us better as humans and make the world a better place. So I’d love to see complexity reduce.
Manish?
I would certainly want enterprises to look beyond the siloed function and doing some incremental stuff, and hopefully adopt the massive total enterprise reinvention. So I would say: embracing total enterprise reinvention.
Coming at it from a talent point of view, I would say make the industry a truly equitable place to be for everybody, where all people from all walks of life, no matter their background, can thrive and add value to the clients they’re serving.
I’d say embrace the digital workforce. Put them to the commoditized part of your stack. Differentiation will percolate up to the edge, where you want to focus your human workforce, which is better at that edge and gives you a competitive differentiation.
I feel this from my point of view in the telecom world — we call it the last-mile problem. Which effectively means: are the benefits of the new technology or the new businesses reaching the last person on earth? Now, that’s really what I would like to wish. That means make it all inclusive.
Make it all inclusive — I think that’s fantastic. Make it equitable, focus the humans on the edges, get out of the silos, make it more simple, think about a generative ecosystem, and think about value and less complexity. So thank you very much, everybody. It’s been a fantastic conversation, and I hope this is the start of a brilliant two days. Thank you.
Your account has been created. You can continue exploring free AI insights while you verify your email. Please check your inbox for the verification link to activate full access.
With the exception of our Horizons reports, most of our research is available for free on our website. Sign up for a free account and start realizing the power of insights now.
Our premium subscription gives enterprise clients access to our complete library of proprietary research, direct access to our industry analysts, and other benefits.
Contact us at [email protected] for more information on premium access.