 Live from San Francisco, it's theCUBE, covering Informatica World 2016. Brought to you by Informatica. Now, here are your hosts, John Furrier and Peter Burris. Hello everyone, welcome back to theCUBE. You're watching SiliconANGLE Media's theCUBE, our flagship program where we go out to the events and extract the signal from the noise. I'm John Furrier with my co-host, Peter Burris, head of research at SiliconANGLE and general manager of Wikibon Research. The next guest is Yelena Roljevic, who's the assistant vice president of Business Intelligence at George Washington University. Welcome to theCUBE, thanks for coming. Thank you for having me. So, obviously you're a practitioner in the trenches, university environment, very fluid, lot of IT, a lot of things going on from student activities to running the university. Share with us the data picture. What's it like there? What's some of the things that you guys got going on there? Okay, so higher education has lots of data. We collect information on our prospects, on our current students, on their life at the university, how they interact with library, how do they come to sport events and games that all helps us understand the engagement of a student and informs any changes in our activities that we have to make to retain those students and make them succeed and become new world leaders. So we have, the way universities have grown organically, you have data in many pockets of those organizations. And I come to higher education from working about 20 years in the government consulting business. So I was actually very surprised to find out how complex universities are. George Washington University in particular being a city school at the heart of DC, you know, it's like running a small municipality. So you can imagine, we have every data type that can be. What a silo that can imagine. And lots of silos, you know, I mean, we have at least three ERPs, one for our financial data, the other that houses our students' information, and then lots of data on spreadsheets and et cetera. So what we've done, what GW have done about 10, 15 years ago, it started to build Enterprise Data Warehouse. But it was very tactical, you know, the reporting information was really at a very tactical level and we needed to get that value from this, from data all the way to our executive leadership and other decision makers across the university. So about three and a half years ago, we started actively engaging with our business functional areas, you know, and change our project management methodology to really being fully agile, you know, we are on two week cycles using Scrum. We bought the blog as our data visualization tool. We already had IBM Cognos in the environment, but we just needed something that's easier for our end user community to use. Our goal is the BI's self service. So my team is working on integrating data from a variety of different systems, ensuring data quality. We also kicked off a data governance program and that's working very well. We tied our data governance program to our business intelligence efforts because business intelligence efforts had momentum and support behind it and to showcase the value of data governance and convince our data stewards to actually support work of data governance. We tied it to BI, so basically our mantra is if it's important and we are asked to put data in the dashboards, in the reports, then it's important data we should maintain and manage. So then we make sure before we publish these solutions that every data element and any assumptions we made are documented in our data governance tool. So you have, I mean, do the students have mobile devices that you provide in the app? So you have data coming from the mobile devices? We have bring your own device, any device you can imagine out there, you know, policy and that is very interesting challenge to manage. I think on average every student brings about six devices that includes in addition to their phone and tablet, you know, and computer, their PlayStation, their, you know, game box, their TV, Netflix. They have all their apps, like Yik Yak and all these crazy apps that are out there. Exactly, exactly. You've been Yik Yak? No. My daughter informs me what's being said about GW whenever she sees something like that. So, you know, we get all kinds of devices on our network and our students are of course digital natives more and more and they really love some of the dashboards that we are putting out there. So the edge of the network is very robust. You have end users using all kinds of apps, very diverse, you bring your own device to work. I'm sure you got a bunch of IOT, but the thing that you guys live every day is the siloed concept of different systems. Now you have remote campuses too, so you have to move across geography or. We have also lots of, we have international campuses. We have campuses in China. We have, you know, in the DC vicinity, several campuses, Foggy Bottom being our main one, but we have our graduate programs in Virginia. We have Mount Vernon campus and quite a few other locations throughout the city. So as you sit back and modernize, so the theme here is modernize your data, data 3.0, all that stuff, all that good stuff. I mean it's messaging, but the impact to you is you have real systems to traverse. Yes. You have good data and move it across different networks and or geography, certainly international. How do you guys approach that? Do you guys get on the white board? Is it pretty obvious? Is it trial and error? You said it's agile. What's the approach? How do you solve that problem? Because that's a really kind of current use case that most people face. Yeah, we do have our enterprise architecture team that helps look into current state and helps us understand what the situation looks like, what devices and applications we already have, how we move data between those applications, lots of point-to-point integrations. That's one thing, it's on my to-do goals to change. You know, using Informatica, we have PowerCenter on-prem and in the cloud and we are working with our Informatica partners on building the true data integration platform. So I was very excited with what I heard this morning at the general session. So we also work very closely with our business functional users. We try to understand their business, their needs for information, which continuously changed because higher education is also highly regulated industry. So you have those pressures, but also we are dealing with new generation. That is every day. The regulation, which is its own challenge, in and of itself, the regulation of the data. But also it's interesting is that you have real consumers. Yes. And this whole consumerization of IT, which we talk a lot about, is real. Most businesses can actually go slower than higher ed, but on the edge of the network, you've got real consumers who want consumer experience. Yes. And you get the highly regulated infrastructure. How do you balance that? What's the, how do you guys do that? It must be really hard. Well, can I add one thing before you ask about that? Because it's, your most important consumers are some of the most aggressive users of these new technologies. I think all businesses have to do this. But one of the things that's interesting about higher ed, as you said it, is that to attract students, you have to demonstrate your affinity for the way that they want to do things and translate the way learning into mechanisms and systems and programs that they find attractive and they immerse themselves in. So let's add that to the answer. And that's definitely a very complex problem. And I think the way we address that, as I said, through enterprise architecture, to talking closely with our business on a regular basis, we have these portfolio meetings. We have governance across the board, trying to prioritize the most important efforts that the business needs to put in place. You know, by trying to, having a very dedicated teams, I think having a stable infrastructure that's stabilized across several past years. And, you know, constantly innovating and constantly looking, going more agile so we can deliver on the top priorities and hopefully get enough out there to support our business with attracting students and recruiting students. We are deploying a CRM system across the board. And keeping students, absolutely. And one of the things my team is focusing on right now is the retention effort and student success rate. I mean, we're building a future leaders of the world and we want to make sure they come to GW, they have some great opportunities there. I mean, for internships, et cetera, but they also have some great opportunities to go other places. Today's students are tomorrow's alumni. Exactly. So as you think about this challenge, you're in the technology organization and it's forced you to partner with a whole range of institutions within the university. How is the relationship, your relationship, to the different parts of the university and those decision makers starting to evolve? I think we have a very good truck record as I said, I spend most of my time working with business. Although I'm physically located, the division of IT is physically located in our Virginia campus, I spend pretty much my entire week in downtown DC talking with head of enrollment, with provost, who is a big fan of data and constantly watching for trends. And it's a very, and I don't want to go because it will take us a couple of days to go into the whole process of recruiting and attracting and getting, building your new coming class and watching these melts because you know that not all students will come and planning, you're at the vim of 17, 18 year olds and you have to plan the housing for those kids, you have to plan, and of course, you have to make the class that you expect to make. So it's a very complex problem and our business constituents need lots of data. Our research, GW just broke into top 100 research institutions, our researchers need information to manage their grants. They're accountable for how they're spending their money and they need the real time information to know what their current budget looks like, how much they spend, what they have to spend, how far are they getting any additional funding to their grants, who's working on their tasks. All of those things in the past were very static and now we were able to get them some cool tools, visual insights into their information that they can use to make informed decisions and then their deans and department chairs also can look at that data and act on that information and they use data for use cases, I have never imagined we can use that. I'm curious, just a kind of side note here, just a tangent on the question is, being an IT and being in technology, you're enabling all this data and they're using data in use cases you've never seen before, are you getting pulled into the whole e-learning concept because now we're hearing that having courseware online isn't e-learning anymore. The students themselves are finding non-linear progressions to data that may or may not reside in the university and does that put pressure on the delivery of the education? Have you been pulled into that? Can you share some insight into some of the dynamics? It's a real problem, people are scratching their heads going, what's a better learning environment, the internet or? There are those debates and you can imagine, I'm learning about higher education, inner workings myself and I'm amazed by questions our faculty is asking and our administrators is asking, so we have a group that's focused on online education, I don't necessarily sit at that table, maybe not yet. But we're definitely looking in better ways to enhance students' experience and it's not just taping a course and providing it online, it's being able to interact with these students and customize their experience and answers you're providing based on types of questions and their skillset and they're bringing to the table for that course, so otherwise you lose lots of students who are probably, who cannot follow the program, which is something in the classroom you can correct. I mean the reality is right around the corner, you've got VR, virtual reality and you've got AI, kind of bots coming, I can see that kind of happening and connecting those dots, but the other thing that's interesting is that you're starting to see the listening, we heard the marketing cloud announcement here around using listening engines and we know that we've reported on our publication that on the admissions side they're looking at their Facebook posts and saying do we want that person? This is the application, they're valedictorian, looking good but also look at their Facebook or Twitter feed or Instagram feed, different data. So the digital footprint now is interesting. Does that come into the mix at all? Absolutely, absolutely. My own daughter was applying for college this year, she's a high school senior and she told me, mom I cannot be on vacation and she said I cannot check my university response because I have to come from the same IP address because in order to predict if the students really going to come, the schools are starting to look at the entire profile of that student and it's tying it to the IP address so they can say this kid came to visit in the college, this kid clicked this many times on the different information. Downloaded a podcast or a video tour. Exactly, so the space is becoming very competitive. There are all these debates out there about higher education, the value of higher education, the cost of higher education. So the university is really using data to try to predict who is really going to come and who is going to fit as well. Yes, I mean one of the things that we are looking in our retention modeling is can I predict how, what kind of experience students will have if they're coming from this high school versus that high school depending did they enroll in engineering or did they enroll in art and sciences and can I immediately offer them more support or maybe some additional classes? You can tell by the metadata if they're really interested. We have kids in high school, I have two in college and two still coming through, Peter's got a daughter and this is interesting, a lot of people say oh, this is a safety school and schools can actually figure out like okay, we know you're mailing this one in but this is the value of the data now. You can actually look at it and the other discussion in the high school is not so much academic being the best and checking off your overseas community service. So they want to fit. Fit is now the discussion in high school for colleges. Don't be so hung up on the D1 and all these different schools. Find the fit. How does that impact you guys? You guys being asked to provide more data sharing? Yeah, definitely, definitely. We need, our admission offices need more data. As I said, we are putting CRM in place because our current ways of interacting with students are failing short, trying to get that whole picture and to understand how good of a fit that student is and is it going to actually show up in summer or is it just going to melt? So I have a question that's a provocative question. So lean back and think about this one. Lean back then lean in. Lean in, lean back, lean in. Think about this. So people who are watching who are interested in higher ed and some of the innovations, what would you share with them? Spend a minute talking about some of the innovations that you guys are doing in higher ed and higher ed in general that you are excited about that could be for a parent or for a student watching that they may not know about. What are some of the exciting things that are happening innovation-wise? Well, I think in GW in particular, some of our research facilities that we are putting in place and equipped with the technology that I have not seen when I went to school and I have an engineering background, I think make me very excited. You know, I get very excited just walking into that building. Our ability to, as you said, customize their experience to their interests and needs because we have information about their interests and needs. I think that's also amazing. It's something we didn't have when I was growing up. You're one of millions people in a lecture hall. Until recently. So those things really excite me, you know. Just the sheer amount of information that we have today on their social interactions that I don't think we are tapping in yet, especially not to the degree I would like to tap in. I mean, I would like to offer incoming freshmen interesting plays at our learner center for performing arts. I want to put the right type of displays in a textile museum, but if we know more about these kids, and I mean, they're amazing by the virtue of being 17 and 18 year olds. I mean, they have all these interests. They will self-organize, they will start to dance. So provide some discovery really for them. But they live in discovery mode. They look at the world. So giving them, you know, looking at our library, for example, who also wants to understand who's coming and what services they're looking in the library so they can improve those services and maybe offer others that they are not offering today. So I think in every aspect we can help students have a better experience, graduate sooner and really become a world leader. Well, it's also interesting that there's been 50, 60, 80 years of research and investigation into how people learn and we still don't know. There's been all these different models of pedagogy. Some people learn by doing, some people learn cognitively, some people learn blah, blah, blah, blah, blah, blah. And over and over they've been proven wrong. And it's going to be interesting to actually have data that actually speaks to how people learn and over the next five, 10 years, have that be translated into real programs that are probably better attuned to how people learn. Fascinating you're at the forefront of that. How are you using Informatica? So we have Informatica Power Center on-premise that is both on-premise and in the cloud that we are using to do our ETL nightly loads. We have a data quality tool that we just recently, this year purchased and installed and we are offering a data quality services to all our constituents at GW profiling data, addressing data quality issues. We are communicating to our businesses how expensive the poor data quality is and I think they're becoming very understanding of those type of issues. We also have Informatica Cloud that we use for data integration across our different systems and to exchange information with external partners. We are looking forward to use Informatica for master data management, which is on our... Two-do list. Two-do list, exactly. So that's basically where we are with Informatica. Don't need a lot of data to choose that one. No, but as I said, we started Data Warehouse GW, was it started a couple, 10, 15 years ago. So I think we are... We did some work, we now understand our data. I think I'm hearing lots of my colleagues that are starting with master data management. I do think that tying it to the real use cases, which is if you really need this data to make decisions, that's probably important data I should care about. That's a great little rule of thumb. Awesome. Well, I got to ask the final question. What got you excited for the conversation here? Anything pop out of the keynotes that you made a mental note of that you're going to focus in on, that's exciting? Absolutely, I attended the keynotes yesterday. That was exciting on multitude of levels, hearing about data 3.0 that Informatica is focusing on. I mean, in hearing this morning and seeing some of the demos on the, you know, the data mapping and those aspects. But what excites me, another thing that excited me yesterday was a conversation between Tableau and Informatica and the partnership that they're having. I mean, both of those are our vendors. So I'm very happy to hear, to see that they're talking and collaborating because that will make hopefully our life easier. Yeah. So if they're coordinating their roadmaps, what I like about Informatica is it's focusing on data. It's really not going in every direction out there that we've seen some other vendors going. They're really focused on creating this central place where you can get the quality information. The value of data. The value of data. And when I look at my team, although we do visualizations, we are also pushing a self-service model. We want our functional users across the university to answer their questions. Business users, easily. That's one of the reasons why we chose Tableau and we have a community of PI practitioners that has about 70 users today. I mean, we have like 4,000 employees so we are not a huge entity. So 70 users is a lot of folks who can go and tap into that well integrated, high quality information and answer their own questions. Having good data governance so they understand what that information is about and how it's defined. And that's where I think the biggest value is and where I think my team provides the most value in working in that space. So our end users can really self-serve themselves and not depend on our queue and in our order of priorities. We still continue to support the broader, larger initiatives but we would prefer to have our housing department do their own visualizations using data. Because they're the customer. Exactly, they're the customer. Yelana, thanks so much for coming on theCUBE and sharing the insights and the data here in the moment in real time. Thank you. At Informatica World. Yelana Roljevic, thank you so much for George Washington University sharing the challenges and opportunities around being real time, consumerization and breaking down the silos. This is theCUBE live in San Francisco. We'll be right back.