 From the Silicon Angle Media Office in Boston, Massachusetts, it's theCUBE. Now, here's your host, Dave Vellante. Digital transformation is the operative watchword today, but what does it mean from a CXO's standpoint? And how do you take those perspectives and bring them into an organization to affect its strategy and turn that strategy into action? Hi everybody, this is Dave Vellante. Welcome to this CUBE Conversations from CXO Perspectives. I'm here with Paul Lewis, who is CTO of Americas from Hitachi, Ventara. Paul, thanks for coming down from Toronto. Thanks very much. I appreciate it. It's always great to be in Boston. Okay, let's start with you and your background. You're a CIO by trade, been with Hitachi and now Hitachi, Ventara for a few years, but tell us about your background. Yeah, so I've been here about five years, running the Office of the CTO, which is a highly vertical-based organization. Prior to that, I was a CIO CTO of a financial services organization for about 17 years. Operating technology, sort of being a practitioner of what it means to create applications and operate IT and implement projects and worry about, you know, the blinking lights in a data center. So it's a very different world being on the manufacturer side, but getting to see different verticals, different industries and applying that has been intellectually appealing. So something I want to come back to, so you were CIO and CTO, which is not uncommon, but oftentimes, CIO is more in a strategy or a pure business role, you had both. So we'll come back to that when we talk about organizational issues, but let's start with digital transformation, as I said at the top. It's the buzzword, you go to every conference, digital transformation, you must not get eaten by your competitors, you must be the disruptor, et cetera, et cetera, but what does digital transformation mean to you as a CTO, CIO, from a customer's perspective? So I see it much more as being, having a customer perspective when you look at your business strategy. So in as much as people say, sort of customer 360 or you're taking a customer-centric approach, it's not really that. It's saying, how do I look at my business and evaluate it from the customer's point of view? So, you know, the three aspects of digital transformation is operational efficiency, new business models, and of course, the new customer experience. So operational efficiency says, you know, if I'm doing a whole bunch of things or to deliver value, a product or service, but the only thing the customer sees is what's on the shelf and what's available to purchase, then everything I do behind the scenes logistics is up for grabs. Maybe it's not as important to me as what's on the shelf, so maybe somebody else can do it, make that efficiency. In terms of new business models, if all my competitors, especially those new digital disruptors have a new way of engaging with the client on the payment, maybe it's a credit card versus cash, you know, capital versus OPEX, maybe I need to diversify my portfolio to be equivalent to that, to find customers that I'm currently not getting, and then finally, new customer experiences. This is the customer point of view to say, the customer wants to buy from you in a certain way, so you better start to sell your products and service in the way to which they want to buy. Just because your product's on the shelf and the customer wants to buy from you online means you have to also be online, and if your customer wants to buy from your competitor, your product should be at your competitor, right? So you've got to think about how the customer buys, not just how you sell. So all that sort of business strategy. So we can poke that a little bit in a positive way. So when you go back to pre-internet days, the brands had all the power. The retail companies knew what the pricing was, you know, the spreads in the stock market were really large, we had NASDAQ on last week at Pentaho World. All we talked about is how they're becoming basically a technology company to sell their services to others, they are transforming digitally. So my point and question to you is, isn't a lot of digital transformation about how you use data to compete and actually maybe regain some of that, you know, market power or at least catch up to where the consumers are, because the consumers today have all the advantage, don't they? Well data certainly is a value producer versus sort of a side effect that it used to be. But it is fair that the consumers have much more buying power than they have before and that's in many ways because of those disruptors. Those disruptors are creating new options for consumers and now consumers have that choice. In fact, the consumerization as a whole has changing how consumers even perceive companies, right? So if I can download an app and if I don't like it now or I can delete that app and download another, they can also choose your product in the same way. They're going to buy your product, they don't like it, they're going to throw it away and buy somebody else's product. They now have the ultimate choice to do anything they wish by from anybody they want to. Locally or globally, the globalization concept is changing the way you need to distribute your products and services too, yeah. So the power actually, an influence has gone to the consumer and it's only data that you can produce and you can consume externally that'll give you that insight to determine where I need to put my puck, right? Where I need to, a hockey analogy, where I need to ensure that I need to have my product in service and available before the customer wants it or they even perceive to want it versus sort of waiting behind the scenes. So the big difference between, let's say being digital versus non-digital is the data. But what does that mean to a CTO and a CIO? So, okay, data, that's the big difference, not what. I would say let's take it from the top. So if the CEO now is focused on creating more value quicker, they probably hire a chief digital officer that's focused on those three pillars. If the organization's not that big, they might have the CIO perform that function. That means the CIO is less about order taking and more about value creation. The only way they're gonna be a value creator is if they move from an application-centric world of IT to a data-centric world of IT. And I use an analogy of applications, infrastructure, and applications. You want me to go through that way? Yeah, absolutely. I'd love to hear more about that. So here's the difference between infrastructure applications and data. If I look at infrastructure at last, let's say three to five years, I might be able to sweat it out any longer, but if I do, I'm gonna have performance, scalability, availability problems. If I add more infrastructure to infrastructure, it's gonna cost me more money. I need more space, I need more power, I need more rack, right? Same kind of true on the application side. If that, the last maybe seven to nine years, maybe sweat it out any longer, I have same performance and scalability problems. If I add more applications to applications, I have modernization and simplification and rationalization problems. And it's not the number of applications that matter, it's that I have the same function point recreated across five to 10 different applications and five to 10 different teams worrying about it. Same cost issue. And data quality issue. Absolutely. But data is in fact the opposite to that. Data is valuable to me from the point that I created to the point that I deleted if I ever deleted. In fact, seeing data change over time is more valuable than seeing it static in its initial state. If I add more data to data, the bigger potential pot of gold I have and the nuggets that I can find, the more precise my algorithms become, the more insightful I'll be able to create from a client's perspective or from product or transaction perspective. In fact, it is the value creator for IT versus the side effect that it's always been. So if you remove the centricity from the CIO from application, which is red, green, yellow projects, to data being the value creator, you start to be a major player in the digital transformation organization instead of sort of being the order taker of projects. So there was a lot of things you said in there that made a lot of sense to me. Let me start with sort of the infrastructure. A lot of CIOs have to spend their time keeping the lights on. And that's not a value producing activity we can agree. There were, and still are, many CIOs that were application centric, as you were saying, and they would add a lot of value through those applications. They had a sharp application development team. They could differentiate through those applications. But increasingly, when I talk to CIOs, you see more SAS coming into play and they're trying to avoid custom modifications. So when I ask them, well, how do you differentiate? The differentiation is the data. The data and the IP that we build around that data, the way that data helps us monetize, whether it's directly or indirectly, is our new differentiator. But that's a big shift, isn't it? It's a large shift because they're completely application centric. All their projects are about versioning of applications, all their infrastructure is creating highly available for applications. So the big shift is say, how do I create an organization that's data centric as a whole? How do I create a chief data officer? And that data officer is elevated to be the peer of, in many ways, the VP of application, the VP of infrastructure. Their organization has all the data centric responsibilities. They have storage and protection and governance and analytics and stewardship. They are the measured by the value they produce for the organization, whether that's operational efficiency or revenue versus the projects to which they deliver on. And that way the output of IT is not just projects, it's not just spend, but it's in fact revenue or profit. Let's talk about the organizational roles. I said I wanted to come back to that and I do, you know the jokes, CIO stands for career is over. I was interviewing John Halamka, who is the CIO of Beth Israel Hospital, a while back at MIT, one of the shows we do. And he was not optimistic about the role of the CIO. He said, hey, it couldn't disappear. And the conversation, it was a CDO conference chief data officer conference. The conversation was, well, CIOs need to pick a path and you've got some experience here. They either have to become CTOs or they have to become chief data officers. Now that was maybe two years ago. I think the narrative has changed a little bit and people have calmed down about that, but you've seen this, these roles emerge, chief data officer, chief digital officer. We just talked about how digital equals data. So I actually see those two roles as, you know, more closely, you know, aligned or not, depending on, and the CIO's role, I think, you know, it becoming more clear as a business and strategy person. But I wonder if you could weigh in as a former CIO slash CTO, current CTO. You talked to a lot of customers. How do you see organizations, you know, what's the right regime? Right regime's not the right, not the proper term, but what's the regimes that you see emerging? I think the big shift determining what those organization roles are from standardization to diversification. So it's less about single provider, single process, single implementation, having a single set of IT services for all the potential workloads and more like what does the business and specific the lending business require and then how am I gonna support that? So it's now I'm gonna have internal services, I'm gonna have a private cloud, I'm gonna have use public cloud offerings, I'm gonna have managed services, I'm gonna go to third party offerings, I'm gonna use a bunch of SaaS, I'm gonna consume a lot of cloud versions of ERP type products and that's the complexity of my environment. And if that's the complexity of my environment, that's the complexity and change of the shift of the roles. The CIO now has to be less about project delivery, in other words, creating applications and more about managing an ecosystem of diverse deployments. They have to manage relationships with public clouds. They have to manage and create business offerings with the CFO and the CEO and the chief corporate officer in terms of creating new acquisitions or mergers, right? The CTO is focused on creating a highly secure framework of delivery so that not only the IT shop can deliver on value but all that shadow IT that's happening out external to create a platform and a secure platform for them to deliver. Because the reality is of every $100 that the CIO has, there's $250 out in the business. So why don't you make a $350 IT budget instead of 101? You do that by providing platforms. And so therefore the CIO is part of the business leader versus being the IT leader. The CTO is looking at platforms and therefore the chief data officer becomes the value producer. They're the one focused almost entirely on creating revenue or creating so much efficiency in the organization that the profit margins dramatically increase. So now business perspective, business perspective, business perspective and everything underlying is ecosystem. It's not everything that I built. It's things that I consume externally. Wow, okay. So again, a lot of things you said in there that make some sense and I want to better understand. So the chief data officer, as you described it, sort of job one for her or him is to understand how to essentially make money with data. Right. And again, I don't want to say go sell your data because that's not always the answer but you're saying you can drive efficiencies. The simplest form, you can cut costs or you can increase revenue. Or you can make better decisions, right? That's the whole champion and challenger concept. You can have a better understanding of your clients or your products and more importantly have a better understanding of clients to which currently don't purchase your products. Right. How do I look at internal information and compare it to external data to say, well, how are those other consumers they're going to other my other disruptors? What are they purchasing and why can't I produce something that's like or at least competitive in that world? So you started off this conversation with three things, the operational efficiency, new business models and the customer experience. So there's certainly the chief data officer, as you just mentioned can affect operational efficiency and ways to cut costs through data. And I guess they touch new business models as well. Hey, if we're going to monetize our data directly or a partner or bring in other data and we talked about NASDAQ before, that's a completely new business model. Or even working with the finance office to say, if I were to make changes to my business, here would be the net financial effect. Right, okay. Now, the customer experience, is that the domain of the chief digital officer really more in that customer facing? Still a combination, but I would agree that the chief digital officer is focusing on creating to matching the selling experience with the buying experience. And that might be new mobile interfaces. This might be creating omni-channel experiences or expanding upon that to say, how do we ensure that we have an integrated channel experience? It's not just that they can bribe a shoe and the website and a shoe in a store. It's that they can go online, look at the shoes, go to the store, have those shoes be brought down automatically as soon as I walked in and then shoes, whether I buy it now, take it home, buy it online, have it delivered to my house before I get home or it's $5 cheaper, five stores down. Right, so that experience will be chief digital officer, but all of that requires data. One can't deliver on all of that unless they have a deep understanding of their products, a deep understanding of how the transactions, a deep understanding of how clients buy. All of that's experience, data-based, whether it's mobile or human created or business data, all combined together. In fact, that's actually a great jump into the sort of the IOT world, the machine or the physical world where I now need to appreciate data that's happening in the store, in the kiosk. And all of that experience data needs to be brought back and combined with the financial data to really appreciate what the transition of that digital experience might be. So those roles do really span your three areas. I can see, just hearing you speak, the chief digital officer might go to the chief data officer and say, hey, I need this data so I can create a customer experience that gives us competitive advantage and I need that data to be accessible of high quality. I maybe need you to pull in some other data points. I need real time, I need it blended. I need integrated with my ERP. Make it so. Exactly, exactly. That can't be too hard, right? And then that involves the CIO to actually provide the infrastructure and whatever, SaaS or internal applications. Exactly, find a means to solve the problem and it's not going to always be built. It's likely going to be consumed. It's likely going to be by. It's likely going to be partner. And so historically it was the application kind of tail wagging the dog. Now it's the data that is really sort of the driver of the bus. Which is why you really need what we refer to as a data strategy for digital transformation. Creating a set of services or capabilities that are focused much more on data than IT. Like we're used to saying IT services. Make sure you have compute and storage and networking available to you. But now it's saying, you know what? You have business data. Let's make sure you have services like store and manage and govern. You have human sets of data. That's blend and correlate and match. And then you have machine data. Well that's much more about grid and point and IOT related correlations. And you need to bring all that together as a series of data servers to which IT provides to the chief digital officer. Okay, you talked about the edge before. How do you see, I mean we're seeing the pendulum now swing back from centralized in a cloud to sort of decentralize this notion of edge to cloud is probably not going to happen. It's going to be some stuff in between. But how do you see, let's follow the data. How do you see and Hitachi and Hitachi Ventara has obviously a perspective on this. You guys are an industrial giant. How, what's Hitachi Ventara's perspective on how the edge will evolve generally but specifically how the data model and the data flow will change? So we see an enterprise information model has having sort of four legs to this table, right? And that one should keep data where it is because sometimes it's physically impossible to move data from where it was created to where it needs to be for analytics. A train is an example and we produce a high speed train. That could be four, five, seven terabytes per day. Well that's almost physically impossible to move to a server to be able to deal with, right? And when you look at larger machines like nuclear power plants and well treatment centers all of a sudden it's almost impossible. So these four legs are, you still need an enterprise data warehouse. You still need a means to collect your business data and produce your thousands of MIS reports. They actually run the business. That is a $10 million machine to which you've created. You then need a content store, an object store because you have all this human unstructured data to which in fairness a good portion of it might be dark. A good portion of it like your 27 versions of your PowerPoint simply won't have any production nuggets of gold, right? But you still have lots of voice and video records and unstructured files that could contain nuggets. Then you have your big data lake where you want to put your information that you want to do perform an analytics on, right? You don't want to worry about the data model. You don't want to worry about how you're structuring the information until you actually do analytics on it. And then finally the edge, keeping data where it is. Have a federated distributed model and only when I want to do and perform specific analytics do I go collect that information, bring it to the core, perform the analytics, produce visualization results. We kind of refer to this as a data refinement mechanism where I'm searching for the appropriate information using those mathematical statistical algorithms in order to create visualizations that we can blend right back into the original sources. So a lot of data will be created at the edge and it'll stay at the edge. And in fact, a lot of data probably won't even be persisted at the edge. It'll be maybe acted on, thrown away and you'll save what you need to save. Is that- Exactly. And you could say that there's going to be data that's at the edge that persist or not. You might have data which might be referred to as the fog where you will collect it at the CO or at the PO, right? And you want it to pop. And you want to be able to perform analytics with a little bit more compute. You might bring some of that data essentially because you want to combine it and blend it with other information. And then you might actually put it into the cloud because you want to combine other organizational related data and do very complex, highly mathematical problem sets. So we almost see it from sort of edge to outcome where there's edge processing, fog processing, core processing and then cloud processing. Okay, so let's unpack that a little bit in the time we have remaining. So you got at least a three, maybe even a four. Maybe it's a three and a three ATR model. Edge, that second tier gateway aggregation point where you're doing some analytics. And then the third tier and I guess maybe the fourth tier, let's call it your own cloud, private cloud or maybe the public cloud where you're doing the heavy modeling and the training of the models. And then maybe you're shipping the model back down to wherever. Because it's now modifying the machine potentially or the machine's understanding of data. And then you're collecting new data based on that new algorithm to which you're now pushing out. All right, we don't have time, but that just totally changes the whole security paradigm as well. And so, well Paul, thanks very much for coming on the cube and having this cube conversation, really excellent work that you're doing. Congratulations and keep it up. Thank you very much. You're welcome. All right, thanks for watching everybody. This is Dave Vellante and this is Cube Conversations. We'll see you next time.