Champions
of Data + AI

データドリブンな革新を推進するリーダー

EPISODE 13

Making the Case for Digital Transformation

Driving any type of change across a large enterprise is hard enough. Now imagine having to transform your organization to be data- and AI-driven so you can improve operational efficiency, accelerate innovation to gain new insights and implement best practices to build data products. John Irvin joins us to discuss this very transformation and the role data and AI play in helping Deloitte achieve net-zero greenhouse gas emissions in the global fight against climate change.

headshot
John Irvin
Chief Data Officer – Information Technology Services at Deloitte
As Chief Data Officer for Information Technology Services, John is accountable for ensuring alignment and adoption of the Deloitte data strategy, standards, and governance as well as identification and implementation of leading-edge technology solutions that support the US data strategy.

In addition to his CDO role, John sponsors Deloitte’s Global IT Innovation technology sensing programs and serves as the Managing Director for the Data and Analytics CoE and the Application Value and Insights CoE. The CoEs provide enterprise-wide platforms, solutions and services that are leveraged for both internal Deloitte operations and client service delivery.

Deloitte refers to one or more of Deloitte Touche Tohmatsu Limited, a UK private company limited by guarantee {“DTTL”), its network of member firms, and their related entities. DTIL and each of its member firms are legally separate and independent entities. DTTL {also referred to as “Deloitte Global”) does not provide services to clients. In the United States, Deloitte refers to one or more of the US member firms of DTTL, their related entities that operate using the “Deloitte” name in the United States and their respective affiliates. Please see www.deloitte.com/about to learn more about our global network of member firms.

Read Interview

Chris D’Agostino (Host):
Welcome everyone. We’re back with another thoughtful and informative episode of the champions of data and AI series. I’m Chris, D’Agostino your host for this episode. DrIving any type of change across a large enterprise is hard enough. Imagine having to transform your organization to be data and AI driven to improve operational efficiency, accelerate innovation, to gain new insights and implement best practices to build data products. In this episode, John Irvin, chief data officer of information technology services at Deloitte joins me to dive into this very transformation. John will also share the role data and AI playing at Deloitte to help them become carbon neutral in the global fight against climate change. Let’s get started, John welcome. It’s great to have you with us on the champions of data and AI. I, Chris, it’s great to be here with you. Yeah, great. Let’s get started. A couple of things we’re going to do sort of to a set of questions here, just going some things just for the audience to get to know you a little bit. And so, one thing is we’d like to think, like what if the day got shortened, if there were only 23 hours in the day, what would you change about your daily routine?

John Irvin (Champion):
There were only 23 hours in a day. My, my first reaction Chris is I would sleep one hour less. But to be quite honest with you, I’m learning as I get older, how important sleep is. So maybe I’d go to bed an hour. Maybe my answer is that go to bed an hour early.

Chris D’Agostino (Host):
Yeah. See, mine is, mine was the same answer. I would sleep an hour less than I would get up and do my fitness routine sooner rather than wait towards the end of the day. So speaking of fitness and things like that in terms of stress relieving, what one activity do you enjoy that helps you de-stress from what I’m sure is a pretty stressful job as the CDO of Deloitte?

John Irvin (Champion):
I do a lot to try to prevent stress. I get up early in the morning and I have my quiet time. I do meditation prayer and even kind of yoga. But when that doesn’t work, I eat.

Chris D’Agostino (Host):
So those seem to be like two opposite ends of the spectrum,

John Irvin (Champion):
They are, but here’s the, but here’s the thing in the last six months I’ve lost 20 pounds.

Chris D’Agostino (Host):
You’re either doing a lot of yoga or you’re doing healthier eating.

John Irvin (Champion):
No, I’m doing more proactive.

Chris D’Agostino (Host):
So, John, what is it about data and AI that keeps you coming back to work each day?

John Irvin (Champion):
The, the fact that I am constantly learning and constantly challenged, it’s really, I think Chris, even kind of the same, a lot of the same ways it’s kept me even at, at, at Deloitte. I, I just learning, I’m surrounded by amazing people, intelligent people, and I’m learning every, every day. And especially in this data and analytics space that, that we’re in, whether it’s, what’s the latest, greatest new capabilities I should be thinking about for my customers. What are regulations that are out there that I need to be worried about? It’s just something new every day. And that kind of keeps me going.

Chris D’Agostino (Host):
Yeah. And it seems like, you know, kind of, again, back to the fact that you’ve been there for so long and you’ve had roles within Deloitte that have been data focused, at least, you know, my career, which is similar duration as yours, you know, we’re, we’re in a really interesting period of time where data and AI and all of, you know, just how organizations become more efficient with data is so such a strong topic, even inside the boardroom, for example. But if you go back 10 years, 15 years, 20 years, that certainly wasn’t the case. Right. And so I’m curious about the culture of Deloitte versus also your own personal path within the organization. Like, did you find yourself sort of navigating, you know, your own career to the data centric projects because you, you sort of could see the future of just how important data was becoming to run a global organization.

John Irvin (Champion):
I listened to, I’ve listened to a number of the champions of data series. You interviewed a lady from Estee Lauder was.

Chris D’Agostino (Host):
Sol Rashidi.

John Irvin (Champion):
Sol Rashidi, thank you. She said something that really stuck with me, Chris, it’s kind of related to this. She said she was to get this right. Uncomfortable with comfort, uncomfortable being comfortable.

Chris D’Agostino (Host):
Yeah. I think, And get comfortable with being uncomfortable. Right. Which is just this idea of don’t stagnate, continue to evolve, continue to challenge yourself. Right.

John Irvin (Champion):
Yeah. And she, she was, well, the part that resonated with me is she kind of said, you know, she’s kind of like that she doesn’t like to get comfortable. And as soon as she’s being like comfortable, she’s got to push herself to get uncomfortable. It doesn’t happen naturally. And that’s what happened to me. I showed up at Deloitte after I got off the road, as an independent consultant showed up at Deloitte running our SAP finance platform. And after a few years, I kind of found myself thinking and no offense, the SAP consultants out there, how many more ways can I figure out how to get journal entries and accounts payable, invoices into the system, or accounts payable, accounts receivable, invoices out. And, and I was just kind of stagnated. I wasn’t feeling challenged. There was a job that opened up on their data warehouse team.

John Irvin (Champion):
And, and I started, I went, talked to them, I exploded more. I said, look, I don’t know anything about data warehousing. But t I have a technology background. I am a software developer. I’ve been an engagement product manager. And oh, by the way, I know all of deletes financial data. Do you want to teach me what I was saying? And they’re like, absolutely come on board. And, and so I got, I do have kind of hands on keyboard, like back into the weeds and to be quite honest with you, Chris, it kind of for a short term and with a short term, maybe like a couple years, it kind of kept me from moving up the ranks in levels, but it set me on a career trajectory that I’ve ended up like in a place, not just career level wise, but passionate and challenged and always something new that I’m very confident that I wouldn’t have got if I stayed where I was.

Chris D’Agostino (Host):
It’s interesting because, you know, in talking with other leaders in companies similar to Deloitte in terms of size, right. Maybe not the specific vertical in terms of consultancy, but the thing that I’m always struck by is there’s oftentimes a lot of fear or worry that the technology will be too difficult for them to grasp or, you know, they’re not going to become an expert in it, or they’re not immediately an expert in it. And the, but the flip side to that is understanding the organization. In my opinion, just having talked to so many leaders is oftentimes so much more powerful than really understanding the inner workings of the technology. Like you’ll, you’ll want to learn those things, but there’s almost no substitute for understanding how the organization operates. What are the, who are the key people who are the key, what are the key systems that you need to be familiar with and just, how does the business run? And so you shed some light there on, like, you brought that knowledge of just basically like kind of the inner workings of the business. And you brought that to a job where maybe the technology you were less familiar with, but how much did that help you to have that insight?

John Irvin (Champion):
It’s been, it’s been huge. And I don’t know what it is about the data and analytics space, but if you look and you start to, you know, again, you watch your series, right? And you start to see just how many varied backgrounds there are. And how many past there were into data in analytics. And I think that’s the truth. And the reality is also, it is such a broad field, right? No, nobody, nobody, nobody. That can be an expert in all of it. And you have to, you can kind of rely though, you’ll you’ll even have specialties. I remember when I first joined the data warehouse team, you were either part of the back end team or the front end team, right? Yeah. I was part of the back end team. I was the data. I was the data guys. I get much more passionate about that.

John Irvin (Champion):
Now I didn’t get the cool, sexy kind of, you know, reporting, which these days it’s be the folks, you know, playing in AI and ML, but I was fascinated by the data problems and kind of being able to but just, just solve problems, understand kind of data, relationships and stuff like that. So part of why I think we see so many kind of very backgrounds, is it such a broad field. And that you can kind of have strengths in so many areas that you could kind of pick your niche even of where you’re like deep in expertise. And then you have just a really good, solid understanding of the breadth. And then if you know how to work with people and manage people and do change management, which is the other huge part of our job, right. You’ll, you’ll be successful.

Chris D’Agostino (Host):
Well, so let’s, you, you talked about change management and you know, that sort of helps us transition into the next topic of discussion here today is, you went about building a successful data enterprise, inside of Deloitte and, you know, a couple of things, like, what did you use as your north star to kind of define success? Like, how did you know you were done quote unquote, or, you know, on the pathway to being done? Like, what were those metrics and what types of things did you look to to signal that you were on the right path? And then, you know, what were some of the key challenges along the way?

John Irvin (Champion):
Okay, well, we’re not done. I don’t know. I’ve met everybody. That’s that that’s, that’s done, but I get what you’re saying. Here’s, here’s the thing it has. It’s changed over the years. Let me, let me, you can help you understand a little bit, and I think I could talk, shed some light for some folks as well, 15 years ago, or so when I joined that the warehouse team, we had a data warehouse, it very traditional business intelligence reports, but we had this mindset of Mr or Mrs. Customer business customer in the firm. Just tell us what you need, and we’ll go build you a dashboard and you come back and three or six months and you’ll be good. We’ll have your, your reports. We had a platform, it was a two letter acronym. I won’t give it out because it’s not that the vendor, that was the challenge. It was our kind of mindset. But this two letter acronym, wasn’t really the people didn’t people refer to kind of our services and our platform more with four letter words.

John Irvin (Champion):
It was not a good situation we had about, you know, a couple of years after I’d been on the team, we had kind of a visionary leader come in and said, this has got to change. We said, we looked at kind of what was most wrong, kind of, not just with our delivery process, but with even capabilities we implement, we modernized our data warehouse. We put in an in-memory platform with real-time replication from all our data sources. People didn’t have to wait overnight for reports and we went and implemented the world’s largest self-service BI platform with Tableau at the time. It’s still pretty big. I don’t know if we’re still the biggest, but at the time I ran the largest Tableau server implementation of any company in the world. And that Chris kind of changed thinking, the challenge there was just kind of getting customers to try to try it.

John Irvin (Champion):
We, the way we did that is we found kind of willing business people and got them on board and let them just kind of be the evangelists of, Hey, this, this team and these capabilities have really changed. Now in that case, we did a, we took an approach of if we build it, they will come. I don’t do that much anymore if stuff’s like so bad. And you think you’ve just got like a no-brainer, maybe you could do that. Like my IT leader basically funded it all. We didn’t go try to align business partners with a bit of value stories and stuff like that. We just did it and luckily it got all the adoption it’s been wildly successful and is all still in place, but that was a driver, but we’ve had kind of, you know, very different things.

John Irvin (Champion):
Five years ago, we really didn’t have a great answer. We had great business. Self-serve BI we didn’t have a great answer for our data scientists, and it’s not, you know, a hundred thousand people there, it’s a couple hundred data scientists in the firm that really needed capabilities. But what we did there is, you know, we, it was still a fairly sizable investment to go enable all these data scientists with a common platform, which by the way, is where we use Databricks on our, what we call our data science lab in that particular case to get the funding. We, we went to all of these, these groups and that were federated across the firm. And we start to put together the business case, and we started shopping it around to, to business leaders who would start to put in money.

John Irvin (Champion):
And we kind of built this consortium of folks that would, would contribute. But really the key thing was we finally just by happenstance knew somebody that happened to work for the Chief Transformation Officer at the time, mentioned it to the chief transformation officer. They said, I got five minutes. Tell me what your idea is. That five minute conversation with one of these, my business, data scientists led to us going down the path and getting seed investments. So that was a case of the challenge there was getting the funding. And the way we did that was kind of partnering with the business to tell a business story. And we just went around trying to sell it.

Chris D’Agostino (Host):
And prior to adopting sort of that Centralized approach, did you just see, you know, a wide range of local solutions with different tech stacks? And how, how challenging was it for you to, you know, work with those teams to transition them from what they might have known and loved, frankly, right to something that was more central and could benefit from some economies of scale, perhaps?

John Irvin (Champion):
Yeah, it was a little challenge, especially with some of our more advanced groups going in different maturity levels of data of teams with data scientists. So, so I think back to who, who is my now my largest user of the data science lab, my finance strategic analytics team, and they already had their kind of servers stood up. They had a little defined team. They use the interesting thing. Um, remember that, that new data warehouse I said, and the, and the self-serve BI, well, we had also made some kind of promises about how they would change their world too. And it, it didn’t really pan out. Okay.

John Irvin (Champion):
So, so they weren’t quite bought into, Hey, here’s the IT guys showing us how they’re going to change the world. So, so they were a really kind of a show me, so we, we did get them along, but here’s the reality. They weren’t the ones they weren’t like banging out and saying, yeah, we’re in this. So it was actually other teams that were even kind of, not even as mature. And so it was really those teams that hadn’t matured yet, who we kind of built the primary coalition around, but we kept the finance strategic analytics team along for the ride. And once they got the platform stood up and they saw a house, kind of a key part of the platform is how quickly they can request and get access to data to study. And that the fact that they have cloud-based like latest and greatest capabilities, they, they were like an easy sale. They’re like, yeah, that’s

Chris D’Agostino (Host):
That, that, that seems to be an important ingredient in, in these transformation success stories, which is you oftentimes hear like maybe more mature parts of the organization around data science, for example. And they’ve had to, you know, we’ve some organizations refer to it as shadow it where they’ve got to basically build up their own environments and they’re separate from the enterprise it group. And then there’s a little bit of a tug of war over that. What seems to be the big, sort of key advantage for the enterprise group is if they can really focus on making the data sets discoverable and consumable and make, you know, really lower the barrier to entry for people to get access to the data. The tooling starts to fall away a little bit. It’s like, Hey, you know, you’ve, you’ve made it so much easier for me to do my job. And, you know, maybe there’s a feature of the tooling that they missed from their own tech stack. But over time I think that, and they focus more on the efficiency of being able to leverage the data.

John Irvin (Champion):
Yeah. And that’s, that’s exactly where we are. We, you know, I started my story saying, Hey, 15 years ago, 12 years ago, five years ago, it’s actually kind of a great segue. It was like, well, what happened to the last, like few years? Two big transformations that were kind of big things. We got the biggest to be quite honest with you is we aligned across all of, at least in the U S right, every Deloitte like, you know, country as a member for, but I’ll talk to us each of our businesses, advisory consulting, audit and tax, we aligned on a holistic us firm-wide data strategy with those four businesses, it and our other kind of shared services. And that has been huge. And when I say we aligned, we, we decided kind of what was important in terms of our strategy. We formalized the chief data officer role.

John Irvin (Champion):
So put leaders in place that had accountability for driving, you know, alignment and adoption to that strategy. And it created kind of what we call enterprise coordination. That’s now happening like light years more than it did previously. And so that’s how that came about. Like what drove that for years, I tried to drive kind of conversations like that from it, you know, like, Hey, we need data discovery and capability and cataloging. And it kinda just got, got nowhere. This all came about from the top down. I mean, like chief executive officer and there’s leaders basically asking, Hey, what’s our data strategy and saying, well, go talk to audit, consulting, tax. And, oh, by the way, go talk to John in IT. Right. And they’re like, is that best practice? And as there is no right. And so that was the first time kind of top down that I think the general awareness of AI and ML capabilities, and everybody’s saying they, they need to be technology companies. And it’s a real easy to understand for executives that you can’t do any of this cool stuff, the sexy finance stuff I talked about without any of the backend, or you can, but, but you’re maybe just the, you know, a day away or an hour away from ending up on the Wall Street Journal for doing something you shouldn’t have because you didn’t really even have your data house. And were

Chris D’Agostino (Host):
One of the things that’s fascinating when we talked, you know, previously, before, you know, before today’s episode was one of the use cases where you’re looking at and it ties into climate change, but it also ties into this compliance issue, which is, you know, you’re running the data platform for Deloitte and you’ve, you’re helping data science teams analyze different data sets to include customer data sets. I’m sure about some of the data that you’re as your own travel data and your own personnel data. And so the, the protection of that information of course is paramount. But can you talk a little bit about how you’re using the travel information associated with the work that the professionals at Deloitte are doing to service customers as part of the engagement model and how that pertains to climate change?

John Irvin (Champion):
With, with world climate? There’s just a whole new, you know, lens, right? I’m kind of looking at what we can do as an organization and as a world, even as individuals to impact, climate change in a positive way. And in fact, Deloitte has recently adopted a strategy called the world climate strategy. If you were to Google Deloitte and world climate, you get a link to a site that talks about kind of our, our point of view. Deloitte has committed to achieving kind of net zero greenhouse emissions by 2030. And now that’s 20 years, even faster than the Paris agreement says, they’d like, right to see that. So that’s a huge deal, but we just talked about this, right. We’re you know, what does that mean? How do we as an organization, right. For just our organization to get to that point.

John Irvin (Champion):
So, you know, our data scientists are actually using the data science lab platform. I mentioned that that uses Databricks to actually try to model out kind of how we actually get there. It’s it is a complex kind of, you know, model to try to think about, it’s not only just kind of what is our revenue growth gonna look like, and what’s that going to mean in terms of head count, but even we have this whole dynamic of about hybrid work environments. And what does that look like? So they’re using the platform to try to answer kind of those more predictive, prescriptive type questions so that our executives can figure out how we’re actually going to reduce our travel to get us to that net zero greenhouse gas by 2030. Cause that, to be quite honest with you is probably one of the biggest chunks that’s gonna, that we have to deal with. The good thing is, I mean, I like this, isn’t that another good thing about just working for Deloitte in general, is, you know, we we’ve said we set really lofty goals. We believe in strategy. We know how to execute, to get there. And, and even to show that we’re, we’re standing up a dashboard, the world climate dashboard, that’s actually going to be transparent for all our folks to actually show that, you know, what that commitment is, but out progressing and how we’re going to get there.

Chris D’Agostino (Host):
And, and yeah, I mean, this is excellent, but in theory, the model development that you’re doing to take sort of an inward look at how Deloitte operates and what is the carbon footprint on a per professional employee basis, right. You know, to include the travel, but also the facilities costs and you get into that hybrid model. And some people are going to be splitting their time between home and an office space. And what does it, what’s the carbon footprint associated with an office in theory, you can take the learnings from that model development and start applying it to some of your customers, some of the clients that you work with to help them on their carbon footprint approach.

John Irvin (Champion):
That’s a common, common theme. Chris we’ve been, we’ve been kind of have an name for it, it’s called like Deloitte on Deloitte. And it happens all over the place. It happens just like you mentioned in articulated probably better than I could, how it would play out, but even in the technology space, right? Like I’m responsible for making technology decisions that are best for, to, to kind of run its business internally. And that’s what drives kind of our decisions. But we obviously also have lots of alliances and partnerships and we go to market, there’s many cases where it’s really advantageous for us to take our learnings internally with technology, even, just for example, with, with Databricks. And I’ve been a customer for several years now, and as our consultants are going out into the market and talking to customers about their challenges and their problems, our consultants leverage the Lloyd on Deloitte kind of learning all the time.

Chris D’Agostino (Host):
Yeah, that’s great. But let’s, let’s shift to, you know, sort of wrapping up today’s discussion really appreciate your time. One of the things that we like to ask leaders such as yourself is what advice they would give someone that’s aspiring to have a career in data and AI. And I think also, you know, the, the, again, tying back to your tenure at Deloitte, because it’s, you know, 20 years of you kind of navigating and, you know, maybe forgoing a step up in a career ladder, you know, sort of position to, to really pursue what you’re passionate about. What advice, you know, looking back, you know, you’re, you’re now the CDO, so you’ve made it pretty far, which is fantastic. But, what advice would you give to someone that’s starting out in their career and, and looks to you as a role model for what they’d like to accomplish?

John Irvin (Champion):
I get asked this question a lot that, for example, there’s a group here in Nashville called the Nashville software school that has kind of students that are in data science program to data analyst programs. And these are people kind of making career changes. I get asked this question a lot. I didn’t hear Chris. Here’s kind of my, kind of top three kind of as now. The first one is the obvious one for these folks is get good data literate. So that’s a more, just a broad kind of advice if you’re going to be working in a space where you want to work in this space. And I was talking with Dr. Paul Barth from click just last week, about data literacy. He’s now the head of data literacy at click. And I like something I saw and the presentation that we were looking at, what does that mean?

John Irvin (Champion):
It means you can, you can read, right. You read and work with data. You can analyze data, but I loved the last part. You can argue with data. So your cause or your, your data literate when, when you’re actually working with it and understanding and using it, but you’re arguing with it. So, give data, literate is one. Another one is kind of more recent is don’t think you have to be a coder, right? Don’t think you have to be a coder these days. I like to say you, you could be a clicker. Okay. You, you don’t have to learn to write Python, right? The tools have advanced that you can develop models without having a, you know, a degree in statistics. So don’t be afraid. Last week, we did a web in our, internally in the firm, focused on our data science lab.

John Irvin (Champion):
It was called auto and M L and the rise of the citizen data scientist. We had 1500 people from 45 countries attend that session. And I, we were just bombarded with people afterwards. And I remember a taxing your manager set up and very interested in analytics, and I’ve done kind of played around. He said, you guys kind of really showed me that I can, I can go do this. I think the transformation that we saw years ago with business intelligence, like with kind of Tableau, I personal point of view, I know we’re not going to, we still need expert data scientists, but I think we’re evolving. So don’t be afraid if you’re not a coder. And finally, last one, it’s probably the most important, learn to tell stories, learn, to tell stories. There’s, I mean, just Google data and storytelling, and you’ll find tons of things, but I can’t under estimate how, how powerful that is if you’re going to have in this space.