Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Phil pods with Accenture’s Chief Strategy and Services Officer Manish Sharma: Same movie, bigger budget? Pressure-testing Accenture's reinvention bet
147 Plays3 days ago
Phil Fersht asked Accenture's Manish Sharma a blunt question: is enterprise reinvention genuinely a different operating model, or is it just transformation 2.0 with a bigger budget
Manish's answer is one of the most candid conversations on the subject we've recorded this year.
In this episode of From the Horse's Mouth, HFS Research CEO and Chief Analyst Phil Fersht and Accenture's Chief Strategy and Services Officer Manish Sharma cut through the noise on what enterprise reinvention actually requires.
They debate the line between transformation programs that end and reinvention capabilities that compound. They name the five things every enterprise has to unblock for AI to deliver any measurable value: process, data, digital core, operating model, and talent. They walk through what the human-and-AI workforce actually looks like inside a services firm at Accenture's scale, including a concrete before-and-after example of how a mid-career Delivery Lead's day has shifted in the last 12 months. And they pressure-test the dominant narrative that AI is a productivity story, arguing the real prize is revenue and growth, but only for the enterprises willing to redeploy the capacity AI frees up.
If you're running an AI program, sitting on a board reviewing one, or trying to figure out why your last five pilots stalled, this is the conversation to listen to.
Connect with Phil on LinkedIn: Phil Fersht | LinkedIn
Learn more about HFS Research: About us - HFS Research
Recommended
Transcript
Introduction to Podcast and Guest
00:00:02
Speaker
you're listening to from the horse's mouth intrepid conversations with phil first
00:00:15
Speaker
hi welcome to the latest edition of from the horse's mouth podcast i'm your host phil first and with me today is none other than a long time friend and uh in the industry i see him as a friend is manis sharma who's uh Been 30 years at Accenture, he ran operations, he ran the Americas and now he's running what's termed as reinvention services, which is Accenture's bet that continuous reinvention is a fundamentally different operating model from a transformation program. So before we get into the substance, I want to ask the blunt question.
00:00:52
Speaker
So. Manish,
Reinvention Services vs Traditional Transformation
00:00:54
Speaker
when you look at what Accenture built versus what you're building now, is this generally a different movie or is it the same movie with a bigger budget and special effects?
00:01:06
Speaker
You really crack me up with that question that you asked. up Very, very clearly it's the former. I think, ah you know, it is a completely different movie.
00:01:17
Speaker
ah Phil, it is my 32nd year. This is the biggest ever change that we have ever done. So absolutely a totally different, you know, ah movie here.
00:01:32
Speaker
I think, and I kind of add a few comments around that, right? One is the traditional transformation. I think it treats change, which is like a project with a start and an end date.
00:01:48
Speaker
And when I define reinvention, right? Reinvention changes how the client operates. It is a permanent capability, which means a modern data foundation.
00:02:04
Speaker
I think redesigned processes and a workforce model where people and AI work together. And I think if a client sees a new capability,
00:02:20
Speaker
that keeps on compounding after the go live, that is reinvention. But if the value stops when the program team leaves,
00:02:32
Speaker
then it is transformation. And I think here we're talking about is reinvention at scale for our clients where the new capability keeps on compounding, you know, after our go live.
00:02:47
Speaker
And that is kind of the true value that we are seeing. Right. so So let's think a bit about this. on
Accenture's Acquisition Strategy
00:02:56
Speaker
You've got your acquisition machine in Full motion recently I've noticed, ah very impressed with faculty for deliver decision intelligence. I've seen you invest in some network capabilities. You've invested in Rectlite as well to develop capability. um But acquisitions in professional services are notoriously hard to integrate and culture each strategy and all that. So what's different about what you're stitching together now? And and um what is the engine and the gaps that you you're trying to fill to really yeah move you as fast as you can forward?
00:03:34
Speaker
So I think first, if you look at historically, I think we are one of the most acquisitive companies in the world. And We have been very, very successful you know in our investments. you know we kind of We measure them in a very rigorous way and we have had phenomenal outcomes you know from this.
00:03:56
Speaker
I think few things on this. right First is how fast we integrate is what matters. I think M&A for us is builds you know beyond what our delivery teams are kind of doing. So it adds.
00:04:12
Speaker
So for us, we look at additive stuff. It augments and it accelerates our time to capability. The goal is for our clients to feel the new delivery shape in days and in weeks and not even in quarters.
00:04:31
Speaker
I think and the integration speed does not only come from accusianss it also comes how we work with the ecosystem so as you've kind of seen some of the, you know, faculty, UCLA, you know, DLB, which is, you know, ah very, very interesting. And it is right now exploding, actually, yeah even as we speak, even in a very short period of time.
AI's Role in Business Operations
00:04:53
Speaker
We have got partnerships with Anthropic, OpenAI, Gemini. We shape how the models operate inside the enterprise, ah you know, workflows with controls, with data bound and also accountability is in place So we, I think, invest in integration the way most firms don't.
00:05:15
Speaker
We have actually dedicated leaders who we pull out of our practice and they run the acquisition and they own the outcome. We also match the integration model to the acquisition. Some get folded into our you know ah core services, others run as a new or an adjacent business.
00:05:36
Speaker
So for example, DLB, we are keeping it very distinct. you know You know, faculty, we are keeping it very distinct. ula We are going to be keeping it very, very distinct, not from because we will use the power of Accenture to kind of help them grow. But they are right now exploding as we speak.
00:05:55
Speaker
and i think some other example that i give like you know look at faculty you mentioned right? Decision intelligence capability it changes how clients run and execute. It simulates trade-offs and really optimizes all the decisions. And it is not just reports, results, you know, out there.
00:06:15
Speaker
Replit, we have invested. I think the debate is not AI native versus traditional software engineers. It is about applying engineering discipline in a new way to build agentic software.
00:06:29
Speaker
And more code generated faster is not the win. The win is that code that translates to business value, which Phil, you have heard, you know, ah you and I have been on the journey on that one, right?
00:06:42
Speaker
Relentless focus on business value. not more at debt not more integration work downstream. DLB, as I mentioned, is another example, which is it adds end-to-end data center capability.
00:06:56
Speaker
It does, you know, it's phenomenal, right? It takes the site selection to deployment to operations so clients can move faster when the AI demands put pressure on the core value chains. So very exciting, you know, and I think now you will see more and more in the coming months.
Challenges in AI Readiness for Enterprises
00:07:14
Speaker
Yeah, I mean, It's incredible to see the speed of the tech, how it's developing. And I think on the enterprise side, they get it in terms of why they need to do this. So they're not stuck on the why, but I see a lot of enterprise customers, ah they need to transform.
00:07:33
Speaker
and They keep telling us they don't know how to pull it off. You know, theyre they're caught in this pilot purgatory trap, et cetera, et cetera. So, um, how do you How do clients generally get ready for Agendic AI? Do you still see a gap where they're two or three years behind readiness or do you think that gap is getting closer now?
00:07:55
Speaker
So I will say here that a lot of companies, they are still treating AI as a tool to deploy
00:08:07
Speaker
and not a change to how the business runs. So if you don't redesign the work itself, the process boundaries, the accountabilities, and for me, one of the most important thing, how does the P&L impact is getting measured?
00:08:26
Speaker
Nothing gets past the first use case, which you and I have seen like probably millions of use cases. So it is not all or nothing. Some parts of the enterprise are kind of ready.
00:08:41
Speaker
The parts where the process is clean, the data is solid, the controls are strong. And you know the some parts of the enterprise are not ready. But I think there is a humongous amount of work.
00:08:55
Speaker
you know And I think I always see that for us, AI has become but a huge tailwind. because just now the company is coming and figuring out you know ah the things, how to deploy AI has become massive.
00:09:07
Speaker
So I think that it's a wrong question to ask if the company is ready for agents. I think the right question will be, where can we put agents today to get real value?
00:09:21
Speaker
And for me, that is around good processes, good data, good controls, yeah you know a good digital core.
Key Areas for Successful AI Deployment
00:09:29
Speaker
That is kind of the real question and some of our clients are scaling now right so it is kind of really moving fast and some are still stuck
00:09:42
Speaker
yeah what do you think gets them unstuck is it trusting the tech more is it empowering people more is it is it building more confidence what what do you think from especially what you've seen in the last six to twelve months are the keys to the kingdom here for clients You know, ah for me, it is about five things that the client should be doing.
00:10:04
Speaker
ah You know, and I think it's always important to kind of keep on remembering that. And that is why, you know, you and I have been chatting this for 20 years and now some of our beliefs are kind of proven now, right? this First one is they have to really focus on the process stuff.
00:10:24
Speaker
You can't automate bad processes and that holds true. You can't AI and identify bad processes. That's the first one, right? Mr. Key Unblock. The second one is, where is your data?
00:10:35
Speaker
And are you really investing in putting a structure around your data? Third one is your digital core, which is kind of really, really important. Is it really out there? The fourth one is your operating model.
00:10:49
Speaker
And the last one I think is, is your talent ready for this change? So I think if you're able to unblock, you know, these five, you're on to something.
00:11:02
Speaker
But you require patience, you require perseverance, you require grit to kind of change, ah you know, ah these five things. And that is why, you know, for, a with I know the clients wants to, and you most of the companies are looking for the sizzle and quick this thing out on this, but it is a lot of hard work. And I always, you know, keep on saying, tell me the line item in the PNL where it is making impact.
00:11:31
Speaker
That has got to be the starting point.
00:11:36
Speaker
Yeah. I'm,
Developing an AI-Native Workforce
00:11:37
Speaker
I mean, this takes me to something, and we we call it the AI velocity gap, which is the gap between individual AI readiness and enterprise AI readiness. And, you know, us as individuals, we're so... empowered to explore new tools, the risks are fairly low, we're willing to take chances. We know we need to really understand them to be successful in our careers. But then some of us go to work on a Monday monday morning and you still have, as you just described, bad process, tribal data.
00:12:14
Speaker
um The enterprises are are not moving fast enough to accommodate individuals. And I worry that um good companies or or companies who haven't fixed their debts are going to lose talent because talent's not going to want to stay in companies that haven't figured this out.
00:12:31
Speaker
um So how do you how do you go about this? I mean, essentially, you've got 800,000 people. um How does change management work in a company your size and how do you translate this to your clients?
00:12:45
Speaker
See, I think on the talent stuff, right, which we have been discussing tons around this, right? I think first for us is a lot of companies are still training people for yesterday's job rather than a world where people and work together.
00:13:06
Speaker
That is why, you know, we are rethinking the rules, the structures, the job family to match where work is headed and not where it has been.
00:13:17
Speaker
So that's kind of point one, right? Second one, which I'll say is, you know, and you would have seen us as you track us well, right? We continue to hire at the entry level and we are building,
00:13:33
Speaker
what I believe is the world's largest AI native workforce. The world's largest AI native workforce. And that is where it is about training for the future that we are talking about.
00:13:50
Speaker
Then I'll give you some example, right? Because it's always good to kind of talk a bit very specific. Like let's take, ah you know, a mid-career delivery lead, you know, in Bangalore.
00:14:02
Speaker
about even 12 months back, a typical way day was spent on status, hands-offs, you know, and some of the ah manual QA. Today, that lead is spending more time on exception management, product thinking and controls, defining what good looks like, right? Validating output, shaping workflows, and coaching teams on new role expectations.
00:14:28
Speaker
And this is a perfect example. This is ei as a growth story I think productivity frees capacity and value only shows up when leaders redeploy that capacity into work that drives growth, better experiences and new outcomes.
00:14:48
Speaker
And without a real deployment, AI fails to deliver. So I think we are on our way to create you know ah probably a scale which has never been seen in terms of creating the world's largest AI native workforce you know as we talk through this revolution, actually.
00:15:07
Speaker
Yeah, and I've i've seen... um You guys are doing a lot of graduate, fresher hiring. um You're changing the sort of nature of your your skills base. Can you talk a bit more about what you see is really working and where the challenges are right now? Like you clearly, you guys are excited about entry level people. What do you think is the most important thing for a mid-career person to consider as as they look at these changes?
00:15:37
Speaker
I would say, you know as the if for us, it is earlier, and you know that is kind of part of our reinvention services, right? Where we are kind of focused on different personas of our clients.
00:15:51
Speaker
But what we require is not a single dimension of skillsets. You can be deep in something, but along with a technology and AI skills, you also need to understand the industry.
00:16:07
Speaker
the process and the data. And this is where we are kind of but you know what we call as reinvention deployed engineers, a very distinct version where we are trying to kind of get in one single the person, a combination of industry process you know and the technology and the AI mindset you know in that person because that is only when you are able to solve the piece. So I think you need to be deep in something, but along with this, you need to build an orientation because every industry is different, every process is different, and you need some deep skills in those in those areas.
00:16:49
Speaker
Interesting. So
Blending Services and Platform Cultures at Accenture
00:16:51
Speaker
how does this change culture? So we we talk a lot about um software versus services, um trying to turn the world's largest consulting firm into something that behaves more like a product company, and especially with these exciting acquisitions that you'll making you're making. manish so But services' DNA is people, it's relationships, it's billable hours. Product DNA is platforms, repeatability, and margin at scale. So those two cultures do not naturally coexist. So what does that tension actually look like inside Accenture today?
00:17:27
Speaker
And how do you navigate it without losing what made the firm great in the first place?
00:17:34
Speaker
You know, a services culture can solve bespoke client problem fast. You just go there, you kind of run, you know, ah align yourself with what the client needs. I think a platform culture uses standardization to scale margin.
00:17:55
Speaker
Right, so that's kind of the two parts. And I think reinvention needs both. So if you look at Accenture, right, the it lives inside the client workflows.
00:18:06
Speaker
It shapes how work runs end to end. You know, the platform sits next to these client workflows. Accenture builds rep repeatable assets, deploys them inside the client workflows and adapt to the unique constraints of an organization.
00:18:26
Speaker
The goal is actually to deliver software like reliability in services with trust intact. Trust is an important part.
00:18:37
Speaker
When done right, And this is what we will you know execute. Platform enables services, diversifies different revenue streams.
00:18:50
Speaker
It gets to outcomes much faster for the client because got to start with the client, right? And it it kind of really, really delivers those business outcomes much faster, right?
00:19:01
Speaker
And it kind of, you know, you diversify to IP data assets and rep repeatable solutions. it leans into what makes each great right. So for us, it is you know a combination because reinvention will require platform culture and ah you know a bespoke solution for a client. So we are going to combine the two. And that is why I feel very good about you know the direction that we are taking.
00:19:30
Speaker
Yeah, it's ah something we've been talking about for a very long time and it feels like it's finally happening where You know, we've got, i think I think the leaders in this market are really trying to think more in a more platform centric way. so you know, when you look inside Accenture with all this AI investment and all this reinvention language, what are the two like KPIs your team actually watches out for? you
KPIs for Measuring Reinvention Success
00:19:55
Speaker
know you know Not the ones in the investor deck, the ones that you look for on a Monday morning. you know What metric do you wish the industry would stop obsessing over entirely?
00:20:05
Speaker
you know For me, the two KPIs, you know they tell whether reinvention compounds are value delivered and reuse like i kind of think about some of the productivity percentages or the vanity metrics, I call them, right? Nobody can attribute do not, right?
00:20:26
Speaker
So the the biggest hurdle is not what KPIs do I choose, but rather that most enterprise don't have an embedded value measurement system.
00:20:38
Speaker
Everyone is measuring after the fact, which is too late to course correct. You know, value measurement has to be built into the workflow is itself, not added at and the end of the quarter, right?
00:20:51
Speaker
So for me, it is about value delivered, which means measurable impact in cycle time, quality, growth, risk, a cost tied to a specific workflow owner. So I'm making it very precise.
00:21:08
Speaker
That is value delivered. Reuse means we do not reinvent the we ah components, the agents, the data patterns and and controls that kind of scale across teams.
00:21:22
Speaker
Right. So for me, those two value delivered and reuse and kind of quickly, you know, ah ah deliver what I call as a speed to value.
00:21:33
Speaker
I love that. In a weird way, value is is what is inherent in people and reuse is what's inherent in software. So that's a really interesting way of pulling that together. so So let's think about the positioning back. So IT services firms face automation risks from the bottom.
00:21:51
Speaker
Consulting firms face AI disruption of the decision layer from the top. So Accenture sits at that intersection at intersection of both serious scale, right? So five years from now, what does success look like? And more importantly, what is the early signal that the strategy didn't work? You know, what keeps you honest about the risk in the bet that you are making, Manish?
00:22:16
Speaker
You know, as I said, 32 years, I've seen the firm grow from, you know, I was like the first few employees when we kind of got hired. I've seen many cycles.
00:22:27
Speaker
Success in five years means client relationships that outlast any single tech cycle. because clients are trusting us to reinvent the core workflows, not just deploy tools.
00:22:44
Speaker
So client relationship that will outlast any single tech cycle. You know, I kind of, I know you kind of mentioned about the disruption. I kind of see that as ah as ah as a tailwind to us because we've got the new stuff and, you know, this kind of sets out for the next almost five to six years yeah a path for all the companies around reinvention.
00:23:09
Speaker
So very, very you know ah focused around ah success in five years, you know ah client relationships, which will last any of these tech cycles that we talk about because you know finally you have to continue to create value for the client.
00:23:25
Speaker
I think an early warning signal because you asked on that is, We win on the tools, but then we fail to produce the outcomes, adoption and the operating model change.
00:23:40
Speaker
So I am very, very focused around we should not be failing on that front. right Because twoies I deployed this and I deployed this model, I did this and all the stuff ah nice stuff.
00:23:52
Speaker
But then I say, where is the outcome? Where is the adoption? Where is the operating model change? So for me, that would be the warning signal failure. Right, right. So strip away the niche, the org chart, the revenue numbers, the analyst coverage, all that
Organizational Change and AI Transformation
00:24:09
Speaker
stuff. And the real work of AI transformation is organizational, as you've said really well. It's not algorithmic. So what's the one thing you are building with reinvention services that you generally get excited about at dinner, right? the The change that makes all the complexity and cultural tension worth it.
00:24:31
Speaker
So yeah it is funny, you know last few days I virtually met five clients and our teams there, right? And I just get inspired every time I meet my teams along with the clients co-creating.
00:24:44
Speaker
And I had you know this big aha moments when I get this really, really you know exciting is the moment when a team, it stops asking, how do we use AI?
00:25:00
Speaker
and start saying, here is the new way work runs.
00:25:08
Speaker
That is important. Get out of it, but talk about the new way the work will run. That's a shift from AI as a tool to AI as a innovation lens.
00:25:24
Speaker
I think it eliminates the process debt instead of preserving it. I think it signals culture plus operating model. That's when the complexity becomes, you know, worth it.
00:25:38
Speaker
And the second one which excites me, Phil, is there's a lot of this thing about ai being a productivity tool. I personally believe, and I think it is our staff that you know our teams believe, AI is the real, real turbocharger for revenue growth.
00:26:02
Speaker
While it will absolutely drive productivity, But the way I'm seeing the AI usage, whether it is in just growing sales, getting your revenue, that is the one that, you know, ah gets me, the the second one that really, really gets me super excited.
00:26:20
Speaker
So changing the way, the new way the work runs. And second, using AI beyond just a productivity tool, but as an absolute sales and a revenue and the growth driver for the companies.
00:26:38
Speaker
Well, that's a great way to finish this conversation because ah getting that conversation away from tools, away from even AI to just how do we run the business now?
Conclusion: The Value in AI Transformation
00:26:50
Speaker
Because these tools are moving so quickly.
00:26:53
Speaker
We probably won't even be calling them AI in another 12, 18 months time, right? Correct. you know And they will get commoditized, but value will never get compromised. It'll never get to those things. Value is what matters in the end.
00:27:08
Speaker
value is what matters in the end. And on that note, Manish, I wanted to thank you for a fantastic conversation. It's great to catch up with you again. And I can't wait for everyone else to share their views on what we've just talked about today. Thank you. Thank you very much.
00:27:25
Speaker
Thanks, Phil. Always enjoy our dialogue.