 Live from Fisherman's Wharf in San Francisco. It's theCUBE, covering IBM Chief Data Officer Strategy Summit, Spring 2017. Brought to you by IBM. Hey, welcome back everybody. Jeff Frick here with theCUBE. We're in Fisherman's Wharf in San Francisco at the IBM Chief Data Officer Strategy Summit, Spring 2017. It's a mouthful, it's 170 people here, all high level CXOs, learning about data, and it's part of an ongoing series that IBM is doing around Chief Data Officers and data. Part of a big initiative with Cognitive and Watson, I'm sure you've heard all about it watching TV, if nothing else, I'm not going to the shows, and we're really excited to have kind of the drivers behind this activity with us today. Also, Peter Burris from Wikibon Chief Strategy Officer, but we've got Caitlin Lepic who's really driving this whole show. She is the Communications and Client Engagement Executive IBM Global Chief Data Office. That's a mouthful, she's got a really big card. And Courtney Abercrombie, who I'm thrilled to see, you've seen her many, many times, I'm sure at the MIT CDO IQ. So she's been playing in this space for a long time. She is a Cognitive and Analytics Offerings Leader, IBM Global Business. So first off, welcome. Thank you, great to be here. Thanks, always a pleasure to be here, so comfortable. I forget you guys aren't just buddies hanging out. So before we jump into it, let's talk about kind of what is this series? Because it's not, you know, World of Watson, it's not Interconnect, it's a much smaller and more intimate event, but you're having a series of them, and in the keynote is a lot of talk about what's coming next, what's coming in October, so I don't know. Let me let you start, because this was, a originally- This was a long time ago. 2014. Yeah, 2014, the role was just starting, and I was tasked with, can we identify and start to build relationships with this new line of business role that's cropping up everywhere? And at that time there were only 50 Chief Data Officers worldwide, and so I- 50. 50. In 2014. And I can tell you that earnestly, because I knew every single one of them, I made it a point of my career over the last three years to get to know every single Chief Data Officer as they took their jobs. I would literally, well, hopefully I'm not a Chief Data Officer stalker, but I basically was like calling them once I'd see them on LinkedIn, or if I saw a press announcement, I would call them up and say, you got a tough job, you know, let me help connect you with each other and share best practices, and before we knew, you know, it became a whole summit. It became, you know, there were so many always asking to be connected to each other and how do we share best practices, and what do you guys know as IBM, because you're always working with different clients on this stuff, so. And Courtney and I first started working in 2014. We wrote IBM's first paper on Chief Data Officers, and at the time there was a lot of skepticism within our organization, why spend the time with data officers? There's other C-suite roles you may want to focus on instead, but we were seeing just the rise of data, external data, unstructured data, a lot of opportunity, rise in the role, and so I think we're seeing it reflected in the numbers. Again, first summit three years ago, 30 participants, we have 170 data executives, clients joining us today and tomorrow, so we're seeing. And six papers later, Caitlin. And six papers later. And we're going strong still. Exactly. Before we jump into the details, this is some of the really top level stuff that, again, you talked about with John and David at MIT CDOIQ in terms of like reporting structure. Now, where does CDO's report? What exactly are they responsible for? You covered some of that earlier in the keynote. I wonder if you can review some of those findings. Sure, I can share that and then have Courtney add. So we find about a third report directly to the CEO, a third report through the CIO's office, a sort of the traditional relationship with CIOs, and then a third in what we see growing quite a bit, or a CXO, so functional or business line function. Originally, traditionally, it was really a spin-off of CIO, a lot of technical folks coming up, and we're seeing more and more of the shift to business expertise, and the focus on making sure we're demonstrating the business impact that these data programs are driving for an organization. Yeah, it kind of started more as a data governance type of role. And so it was born out of IT to some degree, because, but IT was having problems with getting the line of business leaders to come to the table. And we knew that there had to be a shift over to the business leaders to get them to come and share their domain expertise, because as every Chief Data Officer will tell you, you can't have lineage or know anything about all of this great data unless you have the experts who have been sitting there creating all of that data through their processes. And so that's kind of how we came to have this line of business type of function. And Interpol really talked about, in terms of the strategy, if you don't start from the report strategy, yeah, on the keynote, you are really in big risk of the boiling the ocean problem. I mean, you can't just come at it from the data first. You really have to come at it from the business problem. It was interesting. So Interpol was one of our clients as a CDO three times prior to rejoining IBM a year ago. And so Courtney and I have known him. Express Scripts came, yeah. Exactly. We've interviewed him, featured him in our research prior to. So when he joined IBM in December a year ago, his first task was data strategy. And where we see a lot of our clients struggle is they make data strategy in 18 months, 24 month process, getting the strategy mapped out and implemented. We say, you don't have the time for it. You don't have 18 months to come to a data strategy and get by it and get it implemented. Nail something right away. Exactly. Get it in the door, start showing some results right away. You cannot wait or your line of business people will just, you know, they get pulled. But what is a data strategy? Sure. So I can say what we've done internally and then I know you've worked with a lot of clients on what they're building. For us internally, it started with value proposition of the data office. And so we got very clear on what that was. And it was the ability to take internal, external data, structured, unstructured and pull that together. And if I'd summarize it, it's drive to cognitive business and it's infusing cognition across all of our business processes internally. And then we identified all of these use cases that'll help accelerate and the catalyst that will get us there faster. And so client 360, product catalog, et cetera. We took data strategy, got buy in at the highest levels at our organization, senior vice president level. And then once we had that support mandate from the top went to the implementation piece. So it was moving very quickly to specify, you know, for us it's about transforming to cognitive business. That then guides what's critical data and critical use cases for us. So before you answer that, before you get into it. So is a data strategy a means to cognitive or is it an end in itself? I would say it's a, it should be most effective. It's a succinct one page description of how you're going to get to that end. And so we always say- Of cognitive. Exactly, for us it's cognitive. So we always ask very simple question, how is your company going to make money? Not today, what's its monetization strategy for the future? For us it's coming to cognitive business. Have a lot of clients that say, we're product centric. We want to become customer, client centric. You know, that's our key piece there. So it's that key at highest level for us to become a cognitive business. Well, and data strategies are as big or as small as you want them to be, quite frankly. They're better when they have a larger vision, but let's just face it, some companies have a crisis going on and they need to know what's my data strategy to get myself through this crisis and into the next step so that I don't become the person who's cheese moved overnight. Am I giving myself away? Do y'all ever know the cheese? Absolutely, I give this. Every time a new iOS comes up, my wife's like, no, don't do that. But who cares about them? Who cares about the millennials? I do, I love the millennials though. But yeah, so cheese, you know, we don't want your cheese to move overnight. But the reason I ask the question and the reason why I think it's important is because strategy is many things to many people, but anybody who has a view on strategy ultimately concludes that the strategic process is what's important. It's the process of creating consensus amongst planners, executives, financial people about what we're going to do. And so the concept of a data strategy has to be, I presume, is crucial to getting the organization to build a consensus about the role the data's going to play in business. Absolutely. And that is the hardest, that is the hardest job. Everybody thinks of a data officer as being a technical, highly technical person when in fact the best thing you can be as a chief data officer is political. Very, very adept at politics and understanding what drives the business forward and how to bring results that the CEO will get behind and that the C-suite table will get behind. And by politics here you mean influencing others to get on board and participate in this process. Well, even just understanding sometimes leaders of business don't articulate very well in terms of data and analytics, what is it that they actually need to accomplish to get to their end goal? And you find them kind of stammering when it comes to, well, I don't really know how you as Interpol Bhandari can help me, but here's what I've got to do. And it's a crisis usually. I've got to get this done and I've got to make these numbers by this date. How can you help me do that? And that's what the chief data officer kicks into gear and is very creative and actually brings a whole new mindset to the person to understand their business and really dive in and understand, okay, this is how we're going to help you meet that sales number. This is how we're going to help you get the new revenue growth. So in certain respects it's a, there's a business strategy and then you have to resource the business strategy. And the data strategy then is how are we going to use data as a resource to achieve our business strategy? So let me test something. So the way that we at SiliconANGLE Wikibon have defined digital business is that a business, a digital business uses data as an asset to differentially create and keep customers. Right. Does that work for you guys? Yeah, sure. It's focused on, and therefore you can look at a business and say is it more or less digital? Yeah. Based on how, whether it's more or less focused on data as an asset and as a resource that's going to differentiate how its business behaves and what it does for customers. And it goes from the front office all the way through the back. Yes, because it's not just, but that's what create and keep I'm borrowing from Peter Drucker, right? Peter Drucker said the goal of business is to create and keep customers. Yeah, that's right. Absolutely. He included front end and back end. So money, you got to have customers. Exactly. You got to have customers to make money. So data becomes the differentiating asset in a digital business. And increasingly digital is becoming the differentiating approach in all business. I would argue it's not the data. It's because everybody's drowning in data. Exactly. So you use the data and how creative you can be to come up with the methods that you're going to employ. And I'll give you an example. Here's just an example that I've been using with retailers lately. I can look at all kinds of digital exhausts. That's what we call it these days of, let's say you have a personal digital shopping experience that you're creating for these new millennials. We'll go with that example. And because shoppers, you know, because retailers really do need to get more millennials in the door. They're used to their Amazon.coms and their online shopping. So they're trying to get more of them on the door. When you start to combine all of that data that's underlying all of these cool things that you're doing. So personal shopping, thumbs up, thumb down. You like this dress, you like that cut, you like these heels. Yeah, yes, yes or no, yes or no. I'm getting all this rich data that I'm building with my app because you got to be opted in. No violating privacy here. But you're opting in all the way along and we're building and building. And so we even have like, for us, we have this Metropulse retail asset that we use that actually has hyper-local information. So you could, knowing that millennials, like for example, food trucks, we all like food trucks, let's just face it, but millennials really love food trucks. You could even, if you are a retailer, you could even provide a fashion truck directly to their location outside their office with equipped with things that you know they like because you've mined that digital exhaust that's coming off of the personal digital shopping experience and you've understood how they like to pair up what they've got. So you're doing a next best action type of thing where you're cross-selling up-selling and now you bring it into the actual real world for them and you take it straight to them. That's a new experience. That's a new millennial experience for retail but it's how creative you are with all that data because you could have just sat there like before and done nothing about that. You could have just looked at it and said, well let's run some reports, let's look at a dashboard but unless you actually have someone creative enough and usually it's a pairing of data scientists, chief data officers, digital officers, all working together who come up with these great ideas and it's all based if you go back to what my example was that example is how do I create a new experience that will get millennials through my doors or at least get them buying from me in a different way. If you think about that was the goal but how I combined it was data, a digital process and then I put it together in a brand new way to take action on it. That's how you get somewhere. So let me see if I can summarize very quickly and again just as a also test because this is the way we're looking at it as well that there's human beings operate and businesses operate in the analog world. So the first test is to take analog data and turn it into digital data. IOT does that. Otherwise there's no digital in the software. Otherwise there's no digital anything. Yeah, that's right, that's right. And we call it IOT and P, Internet of Things and People because of the people element is so crucial in this process. Then we have analytics, big data that's taking those data streams and turning them into models that have suggestions and predictions about what might be the right way to go about doing things. And then there's these systems of action or what we've been calling systems of enactment but we're going to lose that battle. It's probably going to be called systems of action that then take and transduce that the output of the model back into the real world and that's going to be a combination of digital and physical. And robotic process automation. We won't even introduce that yet. But that's fine. That's going to be in October. Exactly. Look at that one with Dave. But I really like the example that you gave of the fashion truck because people don't look at a truck and say, oh, that's digital business. Right, but it manifested in that. It absolutely is digital business because the data allows you to bring a more personal experience right there at that moment and it's virtually impossible to even conceive of how you can make money doing that unless you're able to intercept that person with that ensemble in a way that makes both parties happy. And wouldn't that be cheaper than having big, huge retail stores? Well, I don't know. Someone's going to take me up on this. Retailers are going to take me up on this. I'm telling you. But I think the other part right is- Right next to the taco truck comes the truck. So you can still- There could be other trucks in that. Cleaner truck and this and that. But one of the things we're going to talk about is you've got to still have a hypothesis, right? I think one of the kind of early false promises of big data in Hadoop and, you know, just that you just throw this stuff in and the answer just comes out. But that just isn't the way. You've got to be creative and you have to have a hypothesis to test. And I'm just curious from your experience how ready are people to take in the external data sources and the unstructured data sources and start to incorporate that in with the proprietary data? Because that's a really important piece of the puzzle. It's very different. They're ready to do it. It depends on who in the business you are working with. So digital offices, marketing offices, merchandising offices, medical offices, they're very interested in how can we do this? But they don't know what they need. They need guidance from a data officer or a data science head or something like this because it's all about the creativity of what can I bring together to actually reach that patient diagnostic that whatever the case may be, the right fashion truck mix or whatever. So does somebody from the Chief Data Office, if you will, get assigned to your assigned to marketing and your assigned to finance and your assigned to sales? I have somebody assigned to us. Well, to put this in kind of a common or modern parlance, there is a design element. You have to have use case design. And how are we going to get better at designing the use cases so we can go off and explore the role that data is going to play? How are we going to combine it with other things? And to your point, and it's a great point, how that turns into a new business activity. And if I can connect two points there, the single biggest question I get from clients is how do you prioritize your use cases? How can you help me select where I'm going to have the biggest impact? And it goes, I think my thing's falling again. Sorry. It's nice and quiet in here. Okay, good. It goes back to what you were saying about data strategy. We say what's your data strategy? What's your overarching mission of the organization? For us, it's becoming cognitive business. So for us, it's selecting projects so we can infuse cognition the quickest way. So client 360, for example. So we'll often say, what's your strategy? And that guides your prioritization. That's the question we get the most. What use case do I select? Where am I going to have the most impact for the business? And that's what you have to work with. But is it the most impact which just sounds scary and you could get an analysis process or what, where can I show some impact the easiest? You're going to delineate. Exactly. Interpol's got his short list and he's got his long list. Here's the long term that we need to be focused on to make sure that we are becoming holistically a cognitive company. So that we can be flexible and agile in this marketplace and respond to all kinds of different situations, whether they're HR and we need more skills and talent. Cause let's face it, we're a technology company who's rapidly evolving to fit with the marketplace. Or whether it's just, you know, good old fashioned we need more consultants. What are we going to do? Yes. I worked my business in. Well, we can go and go and go, but we're running out of time. We've got a full slate. We're just starting. I know. We were just starting these conversations. I haven't thought of a whole other conversation to have. So what should people look for? We haven't even hit the robotics yet, you know? We need to keep going. Good control. Less coffee for us. What should people think about when they think about this series? What should they look forward to? What's the next one for the people that didn't make it here today that they, you know, where should they go on the calendar and book in their calendar? So I'll speak to the summits first. It's great. We do spring in San Francisco. We'll come back, reconvene in Boston in the fall. So that'll be September, October frame. I'm seeing two other trends, which I'm quite excited about. We're also looking at more industry specific CDO summits. So for those of our friends that are in government sectors will be in June 6th and 7th at a government CDO summit in DC. So we're starting to see more of the industry specific as well as global. So we just ran our first in Rio, Brazil for that area. We're working on a South Africa summit. I know, right? And we're seeing, we actually have a CDO here with us that traveled from South Africa from a bank to see our summit here and hoping to take some of that back. Yeah, we have a lot of global. We have a lot of server from Peru and Mexico and Chile. So yeah. We'll continue to do our two flagship North America based summits, but I'm seeing a lot of growth out in our geographies, which is fantastic. And it was interesting too in your keynote, talking about people's requests for more networking time. You know, it is really a sharing of best practices amongst peers and that, you know, cannot be overstated at small events. Community is building. It's building. It really is. We all come in and hug. We all come in and hug. I don't know if you notice, but we're all hugging each other, you know? Everybody likes to hug on the cube. It's like therapy. It's like data therapy. That's what it is. Yep, yep. All right. Well, Caitlin Courtney, again, thanks for having us. Congratulations on a great event and I'm sure it's going to be a super productive day. Thank you so much. Pleasure. All right, Jeff Frick with Peter Burris. You're watching the cube from the IBM Chief Data Officer Summit, Spring 2017, San Francisco. Thanks for watching.