 Live from Cambridge, Massachusetts, it's the Cube at the MIT Chief Data Officer and Information Quality Symposium with hosts Dave Vellante and Jeff Kelly. Welcome back to Cambridge, Massachusetts everybody. This is Dave Vellante with Jeff Kelly and we're here at MIT at the MIT Information Quality Chief Data Officer Forum. Tina Rosario is here. She's the Vice President of Global Operations at SAP, a practitioner really running the data governance operation, right? Welcome. Thank you. Data governance for years, data governance is kind of pushed to the background. Now all of a sudden, all this big data, information quality, CDO, it's hot again. Yeah, yeah. That's got to feel good, right? Yeah, it does. It does. I read something recently about all these companies starting to get Chief Data Officers and there's a big push now to get that role in the company and so it's really nice to see it. Getting that level of visibility. For years already we've talked about data growing so fast, it's such a pain, it's expensive to manage storage, it's so expensive. It's almost like that bit flipped two years ago and data now is the new oil, it's the new source of competitive advantage. So how much of that is real versus kind of bromide? Well, I think that the importance of data at our company is coming to the forefront. We use it in so many facets, from the executive all the way down to the tactical layperson, from getting insight into our customers, to processing our orders and our invoicing, building relationships with our vendors. We consider that to be a top priority for us and over the years have invested in building out this data management capability internally. It had taken a while, I mean I think it was a little tough in the very beginning but once we had that executive sponsorship, that sense of momentum, then it really took off over the past few years. Well, it comes from the top too, so I mean McDermott tells this great story about how SAP ordered like thousands and thousands of iPads when they first came out and Steve Jobs called them and said, why are you doing all these iPads? You're enterprise and ERP and he said, no we are transforming our company, mobile is the future, so how has that affected sort of your role in just data governance generally? Yeah, absolutely. What we see now is it's much more distributed. So it's not just data on premise, on a laptop connected to a network, it's now data on an iPad or it's data on your, in the old days it was a blackberry, right? Or it's data on some device and it's much more distributed and because of that, it requires much more governance because there's more flexibility, there's more freedom where somebody can go and say, hey I think I'm gonna update data from my iPad and oh my fingers are so big, I can't get to the keys. So we certainly find that it becomes much more important for us to have good data governance at the onset and to have systems and processes and tools to make sure that when data comes in that it's clean, that it's not duplicated, that it's at the highest level of quality and in reality if it can't then we go in and enrich it and make sure that it meets our standards before it becomes part of our common database. So do you have a chief data officer or multiple chief data officers at SAP? We don't have that role formally but we do have people in my organization and people at the C level who respect the data position and who give us their sponsorship and support. So you're essentially the CDO, DeFacto is that fair or are there multiple DeFacto CDOs? Yeah I think that my boss Maria Valar is essentially the CDO, the data czar without having that title but she's empowered to run the data management capability for the company. Just out of curiosity what is your title? My title? No, your boss's title. No, she's the global vice president for data governance and management. Okay and she reports into operations. Okay not to the CIO obviously, we have such a discussion here about where the CDO should report it's kind of, I don't want to say it's academic, it's important where the role sits in the organization. Is that changing, is that evolving in terms of the reporting structures? It varies by company but the way I look at it is who cares the most about it? Who's got the most passion? Who can fund it? I think it's a biggie for us because- Follow the money, is this it? Right, exactly, follow the money, follow the passion, follow the political clout. So somebody really has to step up and say, this is important and I'm willing to put my political capital on the table for it, I think that's also important. And somebody who, they don't necessarily need to understand the details but can build the right team. So it can bring in the right people like what we did in FAP is bring in a bunch of folks who had that level of data expertise and let them do what they do best and not get in their way. I wonder if we can just take a quick step back and this is a very basic question. But I think data governance is a term, it's a little bit fuzzy for some people. How would you define data governance? Where are the components that go into a data governance program? Because I think a lot of people, they kind of get it at a high level, okay we're going to make sure these are rules around how data is used. Maybe some rules around, so we make sure we comply with regulations, things like that. But at its heart, what is data governance all about? We try and keep it simple, right, as much as we possibly can. And we consider it to be around four key capabilities. The first capability is having good organization and good practices around data governance, which means rules, standards, policies. And then we look at the right processes, the right engineered processes for simplifying how data is created, updated, and maintained. The third one is we look at data from an ongoing maintenance point of view. So what are the right operations, what are the right tools to automate the maintenance of data because we know it decays. And then the fourth one is having good technical solutions. So having good business driven IT solutions. And so everything that we do centers around those four capabilities. We try and when we talk to folks within SAP, we don't use data speak. So I think even the word governance is kind of a data speed word. We try to drill that down into just regular business language. Like you can say data quality is very conceptual. We try and break that down and say, okay, what is required by the business? What are the most critical bits of information that you need to run your business process? Let's focus on those critical fields, and let's focus on that critical set of information. And that's what we're going to govern. What are the sources of that information? What's the currency of that information? How do we reconcile the differences? And this is the language that you guys use. Exactly. And then ultimately what's the single version of the truth that we're going to rely on that everybody will agree to? That is the right number. Even if it's not right today, we'll make it right. We'll work on making that one right. Exactly. I mean nobody wakes up in the morning and says, gosh, my data is really bad. I've got to focus on that today. But they do wake up in the morning and say, this process is broken. And I can't get the information that I need. It's not trusted. It's not accessible. It's not accurate. So we try to think of it from those points of view and say, what can we do to better enable those business processes to run more efficiently? How can we get the data to them faster, quicker, and with the right level of content? So now, is there a tension between some of the data governance functions that you do and some of the data analytics functions that maybe the business wants to do? Maybe they want to do new types of analysis to look for new insights. And maybe, do you ever have the occasion where your organization will say, well, wait a minute, that's going to cause some issues with the quality of the data or that might not be the most appropriate way to use data, things like that. Is there a tension between the analytics and the governance? No, I mean, I think the opposite. I think we work very closely together. I think it's our job in terms of governance and management to make sure that the data is at the right level of quality and is at the right level of standards. So the analytics people don't have to spend time normalizing, rationalizing, right? So that it's successful to them and there's that handshake there. And so we get requirements from them. They might come to us and say, okay, we're about to run this report. We need this level of data. Can you help make sure that we get it from the right source, that it's at the right level of quality, and that it's available to us? So I think it's a very symbiotic relationship between us. And we need them to help drive our data analytics. So it's not just business analytics, but it's also data analytics. So we look to use tools like SAP's information steward and we use those tools to go in and analyze the current level of data quality. And so we have a good partnership there where they provide us access to those tools and their analytical capabilities. Okay, so what are some of the things that you're working on today? What are the main priorities within your organization? I wonder if you could talk about that a little bit and some of the challenges that you're having executing on those. Yeah. How are you dealing with that? Yeah, I'd say the biggest one for us these days is the cloud. So SAP's vision is to become the cloud company. And so with that comes cloud capabilities internally. And so how do we data govern in the cloud? What does that mean to us? We're in a hybrid environment where we've got some on-premise, some on the cloud as we're moving more and more to the cloud. What does that mean? How do we make sure that we would say we do no harm? We don't want the cloud to come in and harm all the good data quality work that we've done. So how do we work in this new distributed environment where we've got multiple sources of data coming in? We've got the cloud. We've got on-prem. We've got acquisitions. We've got the SAP store online. We've got various ways. Partnerships, yes. So I think it's becoming much more distributed. And some might say that's much more complex. I see it as just an opportunity for us to have more governance and to spread that governance across all these various channels. Well, if you can get it right, then you get it right for your customers, right? Okay, take an example of acquisition. So you acquire success factors. That's in a couple of years now, right? So now, presumably you're taking core SAP HR and success factors, bringing it together. You've got on-prem in the cloud. And you have two totally different data governance models, I presume. Is that a good example of one where you had to dig in? Are you still digging into that? Yeah, I mean, our philosophy for an acquisition has been we all have one governance model. Not too disparate. So they come in and they adopt to our requirements. So when we bring that data in from an acquisition, we make sure that it meets our standards. And frankly, if it doesn't meet our standards, it gets pushed back, right? We draw that hard line that says, you know, you've got to go back and you've got to give us the content that we need. Or it won't come into our central database, right? And then we educate them and teach them what does data governance mean. Because an acquired company may not even know data governance. They may not have had any practices to begin with. So we make sure that they understand what our requirements are, what our processes are, and that they adhere to them. So that's a transition time, right? Yeah, but is it an unfunded mandate for that division or that aquiree, and they have to put engineering resources on that or other resources. And you're not going to pay for it. That's not your role. But you have the authority to basically say, sorry, it doesn't get in here. Okay, so you're the gatekeeper. Yeah, exactly. And we draw a pretty hard line, right? And you get executive support on that, obviously, if you don't, but it doesn't work, right? Yeah, we do, for sure. I mean, we had a painful experience years ago, and we learned a lot from that, right? We learned a lot from that pain and said, never again will we just open the floodgates and say everything comes in. We won't do that, right? We'll take a very structured and very managed, a very disciplined approach to any of the acquisitions. Okay, any other sort of last minute thoughts on advice you would give to other practitioners that are trying to solve some complex problems? Yeah, I would say, and I mentioned about the business speak, try not to use too much data speak. Know your business. Try and reduce the complexities as much as you possibly can. Look for the simplest road, the path of least resistance. Get a lot of friends, we say that quite a lot, right? Get a lot of friends of data. Find those friends in the company that can help support your cause and persevere, right? What I used to say in the three P's, right? Be persuasive with your messaging and business words. Persevere through all the organization changes that you may find, right? Stick to what you know to be the truth. And be diligent, be committed. The third P. I don't know if it's okay, but I know I always say, press on, there you go. But definitely the persuasion and the perseverance, right? Thick skin, I understand. Great. All right, Tina, we'll leave it there. Thanks very much for coming on theCUBE and look forward to seeing you down the road. Thanks. All right, keep it right there, everybody. We'll be back with our next guest. This is theCUBE, we're live from MIT in Cambridge, Massachusetts. We'll be right back.