 Live from the Hilton at Bonnet Creek, Orlando, Florida. Extracting the signal from the noise. It's theCUBE, covering Vision 2015. Brought to you by IBM. And now your hosts, Dave Vellante and Jeff Frick. Welcome back to Vision 2015 everybody. I'm Dave Vellante with Jeff Frick and we're here. This is theCUBE, we're live on the ground. Two days of wall-to-wall coverage of IBM Vision, hanging with the CFOs, performance management, sales management, governance, compliance, risk. John Cothart is here. He is involved in the UI, the user experience around Watson, Watson analytics. The shiny new toy, the secret weapon of IBM. John, welcome to theCUBE. Thank you very much Dave, Jeff, great to be here. Yeah, so tell us about your role. You've got product experience and design in your title. That's kind of cool. It's a great place to be these days. So what does that mean? Well, I definitely have probably one of the funnest jobs for the team. So I straddle a number of different initiatives and really it's trying to bring the pretty pictorials of what we think we imagine the software to be into reality. It's how you do that translation from the pictographs of we think it needs to be designed in this particular way and we think the user is going to want to do this, that and the other. And then how do we build the continuum? So I get to basically play with information sets and ideas and concepts all day long thinking about how is a business user going to get value? And going out to these business users and actually spending time working with them on what are they trying to achieve and make sure that that experience flows from the design elements, from all that prettiness all the way through into something that's usable and tightly harnessed to getting them more analytics and insight. So is your role also all the way from ideation all the way through execution? Right, so I'm the bridge. I'm the bridge from the ideation team all the way into the product management and development teams and try to bring things backwards and forwards wherever possible so that if something's, you know, when we get into the development lifecycle in software you end up realizing that maybe some of those ideations need to be tweaked and adjusted and it's how do we do that in a consistent manner and allow that continuum. So I sort of act a little bit like the glue to bring those teams together and make it happen. So in the keynotes this morning you could see some of the sort of older products and you'd see them and you go, okay, that's nice, the tables and so forth. And then all of a sudden you see this new look and feel and the speakers tell us that now we're going to bring in the Watson experience, wow. And it really shines. So first of all, you must be excited to see that come to light. But what's the essence of what you guys are trying to achieve there? Well, you know, earlier today we talked about analytics for all and maybe we'd take it a step further and talk about citizen data scientists. You know, one of the biggest challenges that Dr. Jing Shear was talking about was there's a shortage of these types of resources that are really remembering their stats courses from university and other. And there's also this penchant for wanting to easily get at information. So how do you take something that's deep and complex like optimization and predictive analytics and make it super simple? Make it as easy for just any user to adopt to. And so it's trying to bridge that gap of, you know, all these users have data. They have information, but they may not have insights and they may not be able to drive those insights into action. And so when we talk about things like the citizen data scientists, it's about trying to get all of that capability into the hands of someone who's probably not seen this type of technology ever before. And it really transcends this need to design around a lot of different use cases as well in terms of mobility and in terms of how people actually work, you know, getting into what they're trying to accomplish and making it simple and straightforward with value. I mean, it's kind of been the Achilles' heel of decision support for decades, decision management. Very powerful, but only a handful of people in the organization can actually leverage the tool set. And it's kind of been a failing, you guys don't use the big data theme, but we do all the time, but it's sort of been a failing of the big data world is that the data really isn't, and the insights aren't in the hands of everyday business users. How is that changing? What is IBM, you know, what are you forecasting in terms of being able to actually achieve that? Well, I think Watson Analytics is the starting point and really it's taking, like you said, we've got these other great capabilities in some of our more long-serving products for us in the business intelligence platform, SPSS and the like. And it's really trying to harmonize and blend the capabilities that people who use those every day, day in and day out with that business user who probably sees them, if you think about forecasting just as an example, people that do forecasting for marketing or human resources or operations, they might do that once a quarter, maybe, maybe once every half year. And it's how do you actually get them to see the insights and then know what to do with that data? So I think Watson Analytics starts creating that environment for them. It creates those line of business professionals, the ability to see that information in a different way and start really engaging in the process and engaging in making that transformation from analytics. And I think when we really look at how we want to transform, it's so that the people that need the information and insight can get at it and they don't need to have a PhD in stats. They don't need to go and be a master data management type person to curate information. And Watson Analytics is starting that for us. It's a journey, though, as Dr. Jing Shear said earlier this morning. We're in a great point where we can create value for individuals today, but there's so much more we want to do. There's so much more we want to extend this out to to really bring that capability to make the citizen data scientists. There was a slide this morning that struck me and I'm trying to cut through sort of the marketing and I wonder if you could put it into perspective. Something to the effect of so that everybody in the room can be as smart as the smartest person. Something along those lines, I'm not getting it exact. And I thought, at first I thought, I got a knee-jerk reaction and I said, well, that can never happen. It's always going to be somebody smarter. But then I said, well, actually, if I've got a helper like Watson giving me the insights, I'm going to elevate the rising tide lifts all ships. So what does that mean? I mean, I'm probably botching the tagline, but what does that mean that everybody in the room is as smart as the smartest person? So I think for us it means that when you're working, you're collaborating. And the whole organization needs to collaborate together and I think what's been the challenge in that whole vacuum of sending things over to the data scientist team and getting them to work on models and getting them to figure out what the right algorithm is and then bringing it back and saying, business, this is what we should do. The idea here is that what we really want to see happen is that you take the hundreds of questions that the smart business people you have working for you, allow them to ask them. Allow them to ask them in natural language. Allow them to have something like Watson actually take that information, take all the plethrough and corpus of data that you've got for your organization. Join it with external data, whether that might be social media data, weather data if you're impacted by that, and actually start bringing all of that information to visual insights. And really what we're trying to do is then, like you said, it's trying to raise that tide, trying to bring everyone up to a level where they can have a more intelligent conversation, right? You talk to a marketer and you say, well, what's a great customer? Who's going to give you the highest customer lifetime value? Well, today they may have a gut feel and in many cases, most of the long serving marketers will have that gut feel pretty much nailed. But what if we could give them 10% incremental insight and make them that much smarter by giving them slightly different visualizations and guided discovery to know what are those other variables that I didn't know about? What's actually impacting my customer lifetime value? If you're an insurer and you start only focusing on premiums and you start focusing on the highest level of premiums, are you missing the boat, right? Could it be that there's a specific population that is a long serving customer of yours and they may only have basic rate coverage? It's those types of insights that we're trying to make really visible across the organization so that when that marketer then has the answer and they can have that conversation with the C level executive, that they can actually all be talking the same language and further when they go to put this in place in action in a decision management system, as you rightly put, which have been lacking, is actually a continuum that allows you to say, here's what we want to really focus on. Now my data scientist, the one that really has spent all that time getting the PhD, can sit there and actually work on something that's more meaningful. It really comes down to time to value as well as that upgrade of people's knowledge and awareness, really of seeing something that really makes sense to their business, to their questions. And I think the natural language interaction, the ability to create the insights dynamically for them and sort of point them to things that they didn't necessarily see in terms of the patterns in the data, I think that's going to be very compelling in bringing and elevating those users up to just a different level of ability to execute for their businesses. So John, how does it actually work? I mean, you're sitting down with an analyst who maybe's never played with one of these tools or even a line of business person. How do you get them started? And what's your kind of experiences to their kind of aha moment? Do they have to have an aha moment? Do they generally find one? Is it the natural language aspect based on their experience with Google and kind of their not work world? How does it actually work on the ground to get that person to try it and then to get some positive feedback so they get into a nice loop? Well, I think we started it out with offering it in a freemium model. We started it out for that reason because we do believe that people will maybe not have an aha moment the first time they use it. We want them to. I mean, we're designing it to try to do that. So what's really happening in the solution and in the service is we're actually interrogating all that data that they bring to the table. And most times a professional has some type of list of information, some type of rectangular, commoner sort of set of data. And when they load that into Watson Analytics, what it's really doing right away is it's unpacking that and it's looking at it and interrogating it. And with our experience of analytics over time, it's applying algorithms to say, what are some of the pieces here that really matter? And so what we hope some of the first aha moments would be is when people see what kind of quality their data is. A lot of times you get some type of feed from some system and you really don't know how good that would be for analysis, whether it's even valid to really look at something like a prediction. And so starting with data quality, starting with the ability to start visualizing starting points where I think everyone comes to the table with a hypothesis, they come with a question in mind, now they can ask it, right? And sometimes that's just the aha that I ask it and the visualization coach says, oh, you mentioned state and you mentioned this, let me show you a map to represent what you're looking for. Or let me show you a tree map or a heat map. And it dynamically does that. It does that without the user having to really know too much about the system. So sometimes that's it. But then you get other users and I'm thinking about Legends Hospitality who's got a publicly available quote for us and their big thing was they were spending months and months trying to find the variables that mattered to a particular business problem. And what they found in a week's time in days of playing with Watson analytics and looking at their data slightly differently than they had before, they got the answer that they were trying to accomplish for the last six months. And so when you start hearing those types of messaging from customers, you start realizing that the all-home moment may not be right away. But definitely in terms of time, we can start creating that all-home moment that much faster for them in trying to get the insight. And so for that basic analyst, someone who's not used these tools outside of maybe a spreadsheet type tool, maybe a visualization tool that helps and build pretty graphics, they can now interact with it. And we can guide them. We can give them that guidance, which I think is starting to make that difference for them. What about? I'm sorry. I always had my visualization question, which I think is funny. I was just going to ask that. A billion of, I'm just wondering how do you visually represent a billion of anything to get a meaningful something on a picture? You know, it's such a great question, guys. And it's something that we continue to work on. I mean, this is part of the journey for us. So when we started down this project, we do a lot of research and development at IBM, as you guys know. We started the visualization engine sort of coach, if you will. We started that project about four years ago. So it's definitely matured a lot in the four years before we brought it to market in Watson Analytics. But what we really do is we're using a series of algorithms and there's actually a whole set of standards that a German body has kind of created to say, what's the best way to display things? What are the right colors? And we're adopting those into our rapidly adapted visualization engine. And what that allows us to do is when someone asks the question and there's break down, we're immediately going to start leaning towards giving you a breakdown, which is really a tree map. You know, when they do compare, I mean, that's when you want to see the side-by-side bar. When you want to see a ratio, right? That might be a donut chart or something where you can sow the percentage of the hole and the different metrics. So we've learned and we've taken all of that learning over the last four years, plus, you know, all the corpus of information that's available to us from standard bodies and whatnot, and pushed it into a smart algorithm. And so it's hard. It's not easy to get it right, but we'd like to think that we're at least giving you a series of relevant opportunities to visualize what you were looking for. And the visualization engine is part of Watson, is that right? Yes, that's right. And that's relatively new for this whole space. Yeah, yeah, exactly. It's not just, you know, an engine that creates pie charts. It's an engine that actually thinks about the data set. It thinks about what the user's asked. And this engine is, you know, it's designed to give you extensibility. So as business users, you can start writing that much more into it and how you want to present your information and create a common way for your users to adopt visualization so that you can actually create standards inside of a business. So I got to ask the competitive question because there's a lot of upstarts, talk to speaking of visualization, that sort of take pot shots at the Cognos and the whole BI world and building cubes and it's slow and cumbersome lacks the visualization. How do you respond to that? Well, I'd say that we're entering into a fairly new market in terms of almost a data-less approach, model-less approach, maybe might even be more accurate. And so for us, the hard part is to make sure that we can carry all those customers that we've got into this new world into a new way of thinking and a new way of managing their information and a new way to know. And I would say that what we're really trying to design is that model by which you can take all the greatness you've created over time and use it, but also be able to work in a much more flexible situation. One thing you'll notice is when you start working with Watson Analytics and you're exploring the data, just as an example, we dynamically create hierarchies where we think they make sense, but we also give you the ability to make the hierarchy for yourself. So you think of retailers and a traditional hierarchy in a more older school technology is likely to put year, then months and days, right, and weeks and everything in that chain. Well, there are lots of times when you actually want to invert that and you want to actually look at the same week or the same month across time, across the years. And in more traditional tools, that's a very rigid and hard thing to do. With Watson Analytics, you can do it on the fly yourself really quickly. And when you ask the questions, we'll actually dynamically flex the model to get there. So I'd say that we have learned a lot about what our customers are trying to do and we're trying to execute that in Watson Analytics to transform the way they're thinking about their data. Try to map the tool to the behavior as opposed to try to map the behavior to the tool. Exactly. What I tell what you call clients, what I call practitioners is if you look at IBM's portfolio, it's kind of better to have overlap than it is to have gaps, right? So there's not a lot of, there's always gaps, but you guys got a big portfolio that's growing. It seems to me that Watson is this glue that lays over that portfolio and allows you to sort of unlock the power through simpler analytics, through visualization, through just more powerful cognitive computing. Is that the right way to think of it that you're unlocking that power or is that integration sort of far off? It's definitely not far off. Some of it's already done and completed and in market, which is great. And I think it is the right way to think about how we want to transform analytics, how we want to make sure that we can get easy analytics, we can get the simple and smart analytics. Because really, when we look at all the wealth of, my good friend Doug Barton who works for us calls it an embarrassment of riches. I mean, when you look at all the capabilities we've got, it really is phenomenal. And what we're trying to do is we're trying to, as we evolve how users are working, as we create a new model for them to be working, we need to make sure that we unlock their potential. And some of that means that we've got to rely on some of these more heritage longer serving products for IBM as we make that transition. So what you see is the ability to right away right now go in and actually connect to your Cognos environment and pull in information and do further exploration. Some of the visual storytelling is actually overlapped very nicely where when you go into business intelligence and you're looking at workspace and then you go and look at assemble, there's a lot of those same concepts of dragging and dropping and asking or doing intent-based offering where you're asking for data to just serve you with the visualization coach. So we are trying to make Watson analytics that forefront for the masses. And as we're doing that, we want to make sure that the people who have invested with us on some of these other tools also get some of that capability. And obviously we're here at Vision and sort of a relatively narrow focus on sales performance management and GRC, but then there's this whole Hadoop space that IBM plays in. That's another corpus of data that Watson analytics can analyze. Exactly, I mean, we at some point, at different points we talk about the styles of data that we're really trying to deal with. Most of the analytics people want to do today in terms of the line of business is definitely in more structure than unstructured. But there is a huge plethora of unstructured data out there. If you look at what we did with the Twitter relationship and why it's in Watson analytics, so that we can actually curate something that's meaningful. Just looking at the tweets over time is probably not as effective as if you could look at the tweets and the surrounding hashtags that were also in that tweet along with sentiment, along with being able to look at some of the demographics of who's actually communicating with those hashtags is important. So we like to think that anything that is related around the big data story, the Hadoop platform, anything that's unstructured, there's immediate value to present it in a slightly different way. We work very closely with the Watson Explorer team and a few other of the Watson teams as well as the big data insights teams to make sure that we can actually have a nice handshake of when we're looking at data in different formats and different styles. So we're out of time, John, but last question. Break out the binoculars or even the telescope. Where do you see this business going in the next five to 10 years? I mean, technology is moving so fast. Cognitive computing is really, you know, coming to real, it's becoming real. Where do you see this business going in the next five or 10 years? Well, I think probably the single biggest thing that I could say is that we are truly going to transform the way that people work and where they work and how they work. And I guess what I'm really trying to get at is form factor does not matter and it hasn't mattered for a while. But now you're getting to the point where you're going to want to be able to do voice search and get answers and insights quickly, immediately. You're going to want to have assist solution that allows you to ask a question and you may not have access to all the information today but one that suggests to you the information that will matter to your analysis. So if you're one of our good customers who talks about us every once in a while, the Cincinnati Zoo who's been up here at Vision a few times for us, talks about the fact that they started with our technology stack and being able to understand that actually having the ice cream bar open at 11 in the morning before lunchtime actually made them just as much money as having it open all afternoon. So they might as well have it open at the right times. But now imagine if you could be factoring in all this other external influencing data, what's happening politically around the region where that particular environment is, what's happening and whether, all of these things. So I think we're going to move into a model where people are truly asking questions of their data and the cognitive computing pieces going to be resolving that question and guiding you to information you didn't know you should have access to. And that's, you know, it's not about the form factor, it's about the corpus of data that we're going to present to you and how you end up using it. IBM is a camelot of analytics for the embarrassment of riches. John, thanks very much for coming on theCUBE. It was really a pleasure having you. That's great, Dave, Jeff. Thank you so much for having us and I hope you enjoy Vision 2015. Absolutely. All right, keep it right there, everybody. We'll be back with our next guest. This is theCUBE, we're live, Vision 2015. Be right back.