 TheCube at Hadoop Summit 2014 is brought to you by Anchor Sponsor, Hortonworks. We do Hadoop. And headline sponsor, WAN Disco. We make Hadoop invincible. Welcome back everyone here live in Silicon Valley for Hadoop Summit 2014. We're actually in San Jose. This is TheCube, our flagship program. We go out to the events and extract the civil noise. I'm John Furrier. I'm joined by Jeff Kelly, a big data analyst at wikibond.org and our next guest is Richard Morris, founder and CDO, Chief Data Officer at Trasata. Welcome to TheCube. No, thank you. It's great to be here. We'd love to have Abhiyana, also your co-founder. You guys are unique in the sense that you're innovative and disrupting and you're pioneering the application market for big data and you're taking a view of not being an infrastructure container but you're really about software around data. So tell us around the founding. Why the founding of Trasata? What was the vision behind it? And why your role as Chief Data Officer so central to that vision? Yeah, sure. I mean, we're one of the key principles. We really believe that data is a key strategic asset of really every company and that the explosion of data and then the new technologies will now go to turn over the entire really infrastructure stacks across all industries. We also had a real clear view that we believe vertical applications were the way to go and really the reason between that, as you see a tremendous amount of variety of data and you need to understand and be an expert in that data on a vertical nature for a specific application to be able to build a credible application. And that's the real kind of one of our core principles and that's the function of our data engineering. Ever since we met Abhiyana and worked with you guys even before you guys started the company we've always been a big vision of have the same vision of this vertical focus because we're in the media business where it's vertical media. So now with the world being instrumented around data you can actually go in and use data to precisely get things and get information and build around that in a way that's unique. It's always been hard in the past. So explain to the folks out there why now, why is big data enabling this really a capability of something we've never seen before? Well I mean, what we do is we build what we call a data index and we do that by the industry verticals or the applications we have and that data index is able to essentially bring the unique aspects of data in that specific application. And so it can get taken, that application can get taken across many, many, many companies but with inside that vertical application. And that's what the key is because if you're in a credit lending business and the type of data and what you're allowed to do with it is completely different than you for in healthcare with the different laws and regulations and type of specifics there. And it's really not credible to think that we as a company could just generically build an application that would be fun to look at genome research versus shall I lend money on a car? I mean as absurd as that is but that's kind of unless you build these vertical applications that's what you're really implying you can do. And so to do that you really need the you've got to have the business domain knowledge to be able to do that. So you know, Abhi talked yesterday on theCUBE about how he's got essentially his head of sales is really a banker at heart and can talk to the financial services institution. How do you go about, I mean what you're trying to do makes a lot of sense holistically but it's a big job bringing together the technical knowledge, the science, the business domain and building these applications that are actually going to move the needle of these companies. How do you go about building a company like that? Well I mean one you've got to pick your verticals correctly. So you need, you know, you can't try and build one size fits all. So pick the specific vertical where you believe that the market size, well first off there's a major problem that hasn't been solved before that really bringing this technology and the data together is going to do something that hasn't been done before. It's not kind of a champion challenger. You know let's do it a little bit faster, cheaper and turn the dial a little bit. Really what are these verticals that are going to completely change over? The business models are going to change as you bring the technology and data together. And then those market sizes are enormous in themselves. So you know where verticals are there where you can build once and sell many, many, many times down that vertical. So you know your title Chief Data Officer. So talk a little bit about that role. You know well what does a Chief Data Officer do in a company like yours and what are you seeing but more generally as a Chief Data Officer at your client organizations? What's the role of the CDO? Well I think we're seeing the client organizations a wide definition of that. But I think, and that's evolving and getting defined depending on what client organization it is but specifically inside Tresada we run a data engineering team which really allows the building of what we call data indexes. And so that's what we build our software product around so that it can go through a vertical application and be sold across many companies inside that same vertical. And it is scalable across that. That's what we see is what we call a data index and that's a very, very cool part of our IP inside our product. So from what you're seeing out in the field and customers and prospects, are you seeing a lot of organizations that actually have created the CDO role? You mentioned it's evolving, it's developing and we've brought theCUBE last summer. We're going to be there again next month at the MIT Chief Data Officer Summit. We talked a little bit about that. Talk to some Chief Data Officers and some of the larger healthcare organizations financial services organizations. But I'm curious from your perspective, are you seeing the CDO role become more prominent or more popular across different verticals? In fact, I think just about all companies we deal with have a CDO. I think the type of person that is in there and their actual role is different and evolving. But I think there's a recognition now uniformly that data is a core strategic asset of any company and it will drive a tremendous change in the business model. And so some CDOs are more governance orientated and facilitators and others are actually building real data assets, some central and some distributed in organizations. And I think because it's new, that's evolving and it depends on the culture of each company. And how do you see the relationship between the CDO and the CIO? Do the CDOs, should they be reporting to the CIO or should they be on par with the CIO? You know, I think, one is I think naturally you have to work in a matrix when you're dealing with this thing. So I think that structure doesn't solve the problem. I think a lot of people go after these things and say, look, if we don't like what's happening and we're frustrated that something's not happening, let's just reorganize the company from a structure point of view. That's just the same chairs getting moved around and you have the same frustration threat. So you need to be able to I think have that role today starts with being able to very much work with Insider Matrix. Some good people are people who actually understand how to get things done and are good executors in a company. So they know that they have the right relationships with the CTO, the CIO, the business leaders. Very important. And they're able to bridge that. But some of that is very person specific. So it's not a very easy prescript of saying, this is great, here's the title. Here's a reporting relationship. Go fix all the data infrastructure problems that have existed over the last 30 years. It's just not realistic. But the people who we see are very good have actually a strong execution background. They might have done a Buzzlet 3 implementation but they have the relationships and the context and they know how to get things done and execute. Therefore, they start out with and they can get something done. They can build a data asset. They can focus on a use case. They can deliver value and get credibility between business and the technology organization and bridge that. And those are the things we see powerful because success, the one thing we really see when you, this is so transformative, the business case is well executed. You're not sitting around deciding was this good or bad. It feeds on itself and said, how do you do more? How does this get implemented more broadly across the organization? But it's so new and organizations and culture and how things get done. I think people are taking different approaches to it. So one of the things we talk about on the Chief Data Officer, Dave Vellante and I and Jeff, always debate is innovation, right? Innovation is going to come from organic growth, seeing new opportunities and big data gives you the freedom and to see new data sets as well as there's a compliance question, right? So old school data governance locked the data down. And so you have this challenge. How do you talk to customers who want to be cutting edge, be agile, super agile, build new things but also maintain compliance because you have control and then you have kind of organic innovation. How do you talk about those two things? You know, that's what their job is. I mean, you've got to be able to, you know, the tensions between those two and that's what the good people, the good executors can manage that. And that differentiates the good from the not so good. I will say one thing then. It's a uniformly, the business models that we see in our clients have to change. So there isn't a discussion of let's keep the existing and add a little bit of innovation around the site. This is really getting into the heart of changing the business model. The margins are going down, people are going online. I mean, the complete change of costs and the pricing points that they're in, they're not making as much money as they used to. That's the imperative to change. As you sit there, the good organizations can manage that, keep tight controls of governance but being able to pick the right use cases and the right implementations of innovation that's getting real results that businesses look and go, oh, that's fantastic. What kind of best practice have you seen that successful folks that do this right? What are some of the things, the patterns that you see, take that big data approach? What are some of the things that you see executing well on the tactical side to manage those matters that successfully? I think they keep their teams nimble. They keep them small. They focus and they're able to build, have very good partnerships and understanding and respect with the business but also with the technology organizations. And you need both, but keep the groups small and nimble, focus on building a data asset first and then build one or two very high use case actual pilots. Then you're going to get, then we see that the order list comes the other way. The business is saying, I went to market with this and here's what I closed. I need this, this and this. And the good ones are actually now taking, filtering the next list of projects that you've got to get done. What are the big challenges that you see? What's the barriers for adopts? Because it sounds like it's an education validation mode. So when you talk about POCs, they want to be conservative because usually certainly in financial services, we talk to Avi about this all the time. Data is a core asset as you mentioned. That's okay. So you don't want to just spill the jewels in the lobby, so to speak and just like screw everything up. People probably take a conservative approach, educate, validate and then expand. Where are we in that life cycle from your perspective? It really depends by client. I mean, we've had our software and production for two years at some places. And to be quite frank, they wouldn't come to this event because it'd be like they're feeling like they've gone back 10 years. Because it's all about tools and piping and infrastructure and they're all around applications and making money and changing the business model. And then other people are just starting the journey and you get the wide spectrum of that. So the folks that don't come to the event, they're looking for the tools to come out of here. They're not in here participating, so to speak. Is that right? And their expectation is those tools are getting built and they look at where the community's going, what have you, but it's not the reason that they're investing in building, right? Are you seeing homegrown has been a big part of, go back to the main frame, I think it's a great example. Spaghetti code, we used to call it, I'm not that generation, but they build their own in-house code. Now you're seeing that kind of same philosophy. Do you see folks building on top of these open source frameworks, building their own homegrown stuff? And what does that do to their business model? They have to obviously hire developers, obviously, right? The homegrown is nearly always taking an existing process and just porting it across. And so that's more the faster cheaper, which is great. Because, look, budgets are going down and even just crimping the growth in the traditional technology, that's great. But it tends to be taking a credit risk process and then just taking it across there. And that's something that's great value when you get in there. But the kind of the thing that's driving the real innovation though is coming in the application from the bringing in the innovation from the application providers. So we'd love to have you kind of size up your competition a little bit. And so the first part of that is, well, what competition? There aren't a lot of big data application vendors out there right now. But the ones that, some of the mega vendors are kind of getting into this market a little bit. You've got IBM, you think maybe Watson and some of the applications are trying to build there. I think SAP was trying to build some of their applications on top of HANA to take advantage of the speed, if not the scale of the data. I mean, well, how do you look at this landscape right now? I mean, you're kind of, you must be a little bit lonely. There aren't really any other startups that are focusing specifically on applications for vertical use cases. How do you kind of look at this landscape? Well, it's exactly right. I mean, we were, Avi and I both sit and look at each other and go, what are we missing here? Where the really aren't application companies and you know, Mike Olson I think two years ago said, it's coming, it's coming, it's coming. And I think cruelly now, I think the next 18 months we will see a lot more applications come in there. And I know everyone laughs when people see that, but no, there are some days where you wonder whether we're wrong and everyone else is right and we should be building, you know, widgets for infrastructure. But no, I don't think we've ever wanted that. Well, what do you think about those mega vendors? The application vendors of yesterday, can they move into the future and today and tomorrow with the IBMs of the world SAP or can they transform their approach to applications to fit this big data paradigm, do you think? You know, I mean, that's the big question. I mean, my guess is some will and some won't. I think they will look very different in five years than they do today. The great companies will be able to transform and do that. I think the biggest thing that they, tremendously difficult to turn a mega ship, you know, five degrees. It's just very hard to do that. Having said that, the market is so enormous that they have no choice and it will become, I think the sole focus of what they have to do. The good ones that are very driven that have the good leadership, you know, and then it's not hard to guess who those are. You know, we'll do that and they'll look very differently and be actually probably better companies than they are today. And others, you know, will go the way that we've seen many of big providers go over the last 30 years. Yeah, another topic that John and I briefly touched on in our intro earlier was the cloud and delivering applications as a service in the big data space. We haven't seen a lot of talk at this show about the use of cloud, but you know, in our data that we're getting back from the Wikibon community, there's, you know, 56% of respondents to our recent survey said they're using big data, sorry, the cloud in some aspect of their big data deployment. What's your approach at Trasata to delivering either as a service or delivering on premise? Do you have a, do you do both? What's your approach and your thoughts just on the value of the cloud as it applies to big data? We do both, but it's a private cloud. But yeah, we know we have, you know, a major global financial institution essentially has a private cloud. It's completely within their firewall, but you know, they have it as a service, fully as a service. And I think that's a very powerful model, to be honest, because I think these companies have to focus on, they've got so much to change, they have to focus on what their real core asset is, they have to focus on and, you know, get better providers focusing on what they're good at. And that's when we talk about how do these large enterprise players behind us change. Our large clients are having to do exactly the same thing. And so we're able to focus their efforts where they really differentiate with their clients and get providers to do other things. It's just part of what they have to do. We, yeah, I mean, the reason I asked that is because it seems to me there's a similar benefit to kind of the approach that your state is taking that generally speaking, the cloud can provide. And that is abstracting away some of that complexity of all the technology and tools underneath, delivering some value in the form of an application and letting your client organizations focus on their business and what they do best. Not be a Hadoop shop, be a bank, be a retailer. And we have exactly that, largest, most complex companies. What they will not do is chip it out to a public cloud. It's just not going to happen. We don't even see, I'm looking back, I'm not sure there's ever been even a chink to even have a discussion on that. But having said that, you described a very different model than what a lot of these large companies have been built on, even with a private cloud. It's a very different model to do exactly what you said. And we absolutely see that. So I was just talking about this compliance thing. We're getting some questions. So the one is that can you give me a specific example of a verticals where it's different, where you guys have been able to provide a solution for where like say with healthcare versus financial, where you guys have delivered value with Triseta's approach and how easy has it been to move from different verticals to different verticals. So the question coming from CrowdChat mainly is around how easy is it to move from verticals? You need domain expertise. Absolutely. So can you describe how you guys move from vertical to vertical? Exactly, so we started financial services which itself is enormous, is a whole series of verticals. So even with inside that between the consumer space to the commercial space to the wealth management space and with inside that we sub-segment that. But there's a common there as financial data there's a common element across all of that. And so we are able to have domain knowledge that it's reasonably transferable across that. Interesting, we have I think about so three months ago moved into retailing in grocery. And again, we were able to leverage our knowledge of financial data in the grocery space as a tremendous, a lot of transactional, it's transactional data, a lot of it is transactional data in nature. So you'd go to Jason Marcus, where you had some leverage, so obviously financial data mean credit cards, obviously purchasing. Exactly, everyone, not everyone, but the vast, vast majority of people will purchase like credit card, debit cards, or. So Avi Hopper, I had a set on theCUBE yesterday that just say there's going to be a disruptive and they're going to disrupt how enterprise software gets built and deployed. What does he mean by that? Well, you know, I usually listen to his and he's going to try and guess what's going on. Well, it's one of the things that we truly, Avi and I truly was one of our core tenants when we started is we really, as buyers of some of that, just could not, it really was just such a disappointing experience. But when we think about vertical nature and now it's data, you have to have an inherently domain expertise to be credible in it. It's not the old way of the old salesperson who said, would you like a cup of coffee and I'll give you some tickets to the football. That is just not useful at all. You have to be able to go in and be very credible and talk to somebody about what their pain points are, what they understand about compliance, what regulatory controls, governance controls, and what really, and to be build a scalable product that incorporates that. Because we want to build once in a cell many, many, many times inside those verticals. A traditional sales model, an info sales model is quite frankly worthless. You guys are very impressive. We've been very impressed with what you guys are doing. Because you're doing the app side, which is not so much the infrastructure and the containers and the storage and all what not. So we have some other good friends that I've interviewed at Factual, Gil Albez and Tyler Bell, you run this product and Gil was the inventor of AdSense from Google. They're building out a big data company, but they talk about this notion of all the folks building technology really don't have any experience with data. They're not data guys, they're like infrastructure, containers and tooling. When you got companies that are data full, they have data all the time, so their view is different. Can you talk about the dynamics? You kind of hinted at earlier that the folks that you sell to don't necessarily come to the show because they're not in the infrastructure business, but a lot of folks here, they're not data guys. They're really not living and breathing massive data. They're infrastructure guys. Yeah, and so I think there's two things on that. One is we were having this conversation a couple of nights ago, and someone who was a very low respected person in the West, right here said, there aren't data people here. And so unless they're the obvious places that have tons of data, but not in our space, there's just no data people here. And there's a reason why we built the company on the East Coast is that's where the data expertise is. Certainly financial services, they're data. They're like fat with data. I mean they're like explode with data. Yeah, and I mean that's all they are. There's nothing else. I mean my daughter when I worked in banking dropped me off at the office and she goes, well what are you doing there? And it was like, hey, well that's kind of a tricky question. You know, electrons moving around, I mean that's it. Well Rob Bearden clearly said on theCUBE yesterday, absolutely the tsunami of data is coming into the enterprise at unprecedented levels and Jeff's documented that in his research. So I got to ask you, where are we in the innings? We like to use the baseball analogy since baseball's in season and the Bruins are out of the Stanley Cups. We can't use the hockey analogy, skating through where the puck is. But baseball, what inning are we in right now? Are we with respect to the data officer and this notion that data is a programmable asset or workable as a key asset? Is it still really, really early? I mean as the game has begun, is it where are we in that spectrum of? I think we're in a very exciting time because I think there are some companies, and I know directly our clients there, that are actually quite deep into the game. Which means they've dealt with these basic platform issues several years ago, they're out. They have delivered real results and these are in production and this is changing their business process. It's being automated and it's not a cost play. Yeah it is cheaper and faster but they are winning business versus their competitors. They're driving revenue. They are driving revenue versus their competitors. They're increasing share and they're documenting it. So those players, I would consider a deep in here. Now they're taking that now across other use cases. They're starting to change their business model. But I would say there are others who are, they're stretching. Well that's consistent with Jeff Kelly's survey. They're just stretching. Jeff your survey had specifically three questions. Big data drivers, obviously drive cost reduction. I was like 20, 21% or so exclusively. And then one on the competitive advantage front which we were referring to was like 24%. But the both was 54% roughly in that neighborhood. Jeff, Jeff you agreeing with that? Yeah I think you're either doing this to save money, make money or both. And if you're only doing it to save money, you're missing a huge opportunity. And we see a lot of practitioners, especially the ones who are putting in Hadoop themselves and kind of becoming a Hadoop shop. They say why? Well I can bring in Hadoop, I can offload some more clothes from my data warehouse and I can save some money and then hopefully I can use that savings to fund some actual innovative work that's going to make me some money. So you're kind of using that saving money to get started but ultimately if that's all you're doing you're not taking advantage of these capabilities. It's about making, it's about creating new lines of business, new ways of doing business. And if you're not taking, if that's not your goal. And incorporating you data, data sets into your business model. Yeah if you're just trying to do it faster, a little cheaper, I think you're missing a huge opportunity. We had a product announcement today that was a 3.3 version. That's pulling in a tremendous rich data sets at an individual entity level. And being able to make and model and get into your application, data that never existed, it was never bought in before is a huge advantage and that's not driving cost down. It's extremely cost efficient to do that and it's fast. But you're doing that to compete in the marketplace. Yeah and I think there does need to be more education around. I think people, there's a mindset change that needs to happen and people are so used to saying well this is the day it's available to me and that's all I have available. The idea that you're limited now is gone. You've got to change your mindset. Absolutely, you assume storage is free, it computes infinite, what problem would you solve? What business would you do? You get in that mindset and by the way you talk about who differentiates, they're the clients that get it. They're the ones who are getting this thing. They're not. I've got to save X million and my target. Let's go beat down a few vendors and move something across there. That's not gonna win. Richard Boris. But I would just like to say one thing. I think we're really doing ourselves a disservice in the community and the people behind us here. When we represent Hadoop as a storage platform and we're trying to kind of, not we, but kind of tread carefully around the different players that we can all live together and it's a storage platform. It is and I don't think we do ourselves a favor of doing that. It is not and it's. It's not just about storage. It's not. It's actually not. And that's why you're going to talk about this just to show being data full and being with data as an application of the buyers. Exactly, exactly. We're here with Richard Morris, the co-founder of Trisada, chief data officer. I'll give you the final words. Share the folks in your own words right now. The future. How do you see the future evolving with respect to how data will infiltrate and change those business models? What things will materialize? What will be some of those new experiences, new expectations, obviously competitive advantage you touch on as critical and still early days. Some are deep in the game, 20 some percent. So huge numbers to drive that forward. What's the future look like for the folks out there? I think that the amount of the use cases are infinite. And so I think that the most creative companies are going to build a data asset and they're going to be able to innovate across that and the applications across that that will be able to build unique companies that you see it like, you know, we're in Abhinaya with Airbnb this week. There are companies there that are going to do banking. There are companies that are going to do capital markets. So the companies there that have not been generated, they don't exist today. And they're the people who we're going to be using in five years. So you see the disruption coming out of the woodwork. There's no question about it. Data has a competitive advantage and opportunity for startups too. One application could explode. They will, absolutely is going to. So we're at the dawn of the beginning of the changing of the guard in your mind. Absolutely, no question. New companies, new value, big data, center of it. It's not just about the infrastructure and the story, just about what you do with it. Richard, thanks for coming on theCUBE. This is theCUBE, our flagship program. We go out to the events, instruct the civil noise. We're here with the co-founder of Trasada. We'll be right back with our next guest after this short break.