 Okay, we're back here live in Las Vegas for IBM's information on demand. The hashtag is IBM IOD. And if you want to participate in the conversation, go to crowdchat.net slash IBM IOD. We'll be documenting all the epic tweets and posts on LinkedIn and Twitter. I'm John Furrier, the founder of SiliconANGEL, my co-host, Dave Vellante. And our next guest on theCUBE is Harriet Freiman, the director of big data and analytics at IBM. Welcome to theCUBE. Thank you very much. So, honestly, we just had a great chat with Tim Buckman where he's talking about just many use cases of the business model of delivering health care, but a lot of the operational. Some of his epic sound bites are situational awareness, data in motion, data flow patterns. These are the insights. So talk about IBM's positioning here today. Last year we were here, big data was being integrated in, IBM's all these assets. Now you guys seem very unified around those key areas of data analytics and social business. While under the hood you've got cloud and mobile, which is the engine providing it all. What's the key thoughts here for the folks to take away? Yeah, sure. So we really have looked at the four technology disruptors in the market today. The social move, which is really a new way that people are gonna be interacting and making decisions. We have the mobile, so everybody's mobile. We need to take advantage of understanding where they are in order to use analytics to better serve them in their location. As well as that being another great source of data. And then we have cloud. The ability to deliver analytics, big data, wherever people are, data born on the cloud. And really what we see with big data in analytics is our fourth area is, how can we put all of that to good use? For example, as you said, in different industries, health care, retail, industrial, et cetera, so that we can really help people get the insights they need and help them apply it where it really matters in their business. Talk a little bit about the relationship between those four pillars. How do you look at the interrelationship? Yep, we see them very much as fueling each other. So where social can be a great source of information, what people are thinking, what people are talking about, what their preferences are, what their sentiments are, that's a great source of information that can help fuel and augment what enterprises have inside their enterprise. So we see social as a fuel for big data in analytics, as well as big data in analytics fueling that social interaction. So very much there's an interweaving in each of those four areas with each other. So yeah, the social, now help me understand your thinking on this. Social data is some kind of fuzzy, right? Sentiment, I'm inferring. How do you go from that level of interaction to something that you can put more confidence in? Talk about how IBM approaches that. Sure, well, we have many organizations that are using analytics on their enterprise data to look at sales trends, look at financial status, the cost and profit equations, using really enterprise structured data. And where social data comes in is it really helps people understand what's happening in the world around them. You've heard the phrase, the world is getting instrumented, right? So we ourselves are getting instrumented in that everything we say we tweet about is a source of information. And if we can look at that and tie that information to the structured data, we get new insights. So let me give you an example. I may be a brand manager and I'm looking at my product sales with my internal ERP data and it's not looking good, let's say. And I go, well, why are those product sales in trouble? I could look internally at my manufacturing, my inventory levels, all the structured information I used to look at. But really what may be happening is there's another brand out there, another product that's competing with me that people are chatting about. They have great sentiment about that product. And actually that's hurting my product sales. So if I can get my arms around what people are saying in the marketplace, understand the text, whether it's positive or negative, the sentiment, what people are talking about. And I can tie that to my internal data. I can now work out the levers I need to have in marketing and sales and service to build my reputation and be able to tie what people are saying about me in the marketplace to what's happening internally. Or make an offer. Yeah, exactly. Make a real time offer. That'd be great. Harriet, Dave and I coined the term crowd consumer and crowd consumer is what we've been talking about in our research around this new consumer category that's out there. That's on the go. You just mentioned the world around. This is a new phenomenon. This is not like, it hasn't really, I mean there's always been people out there. It's always been kind of out of band, half your advertising, you don't even know what you're spending it on as the old adage goes. But now with instrumentation of social data, social business, that analytics is very important. So I have to ask you one, this new consumer is now part of the activities. They're either buying directly, they're participating with sentiment so you can listen to the customer. I mean you should actually listen to the customer for the first time in the history of the world, right? Yeah. Directly. So how does that change from a product standpoint looking back at the historical data world? Business intelligence data warehousing have been slower, mostly fenced out organizations, stored on big tape backup. Now that data needs to be real time. So explain to the folks what's different about big data analytics in the impact of the business owner? Yeah. The enterprise, the business owner saying I want value. I think every business owner and leader that I talk to in my discussions really know that there's value in big data but they don't know where that value is or how to find it. And so we really see a shift happening both in the big data side of things and in the analytics side of things. So in analytics people typically frame of reference is that historical reporting as you said a way to sort of look back in time and sort of have digest the information more as a strategy and management process. And really if we're gonna take advantage of that data community out there we need to think more dynamically. We need to think first in real time how can I understand what's happening right now to change the outcome before it sort of trends down or trends away I don't want it to. So definitely you talked about the sign about real time. We see where big data analytics comes together the most is in real time and that's why we've invested in infosphere streams. That's why we've invested in predictive analytics in those streams so that we can say and we can monitor when something's happening in real time we can assess it. We can predict what the outcome might be and we can suggest the next best action a coffee coupon like you suggested next best action which is gonna have a direct positive impact on the business versus a month from now going oh I wish we had sort of solved those customer churns a month ago. Tim was mentioning about his old, his dad used to take him on Sunday drives and that was his analogy using and essentially what he said was they wanted to go from point A to point B they kind of knew they wanted to do that but there's a lot of different choices to make and most businesses say I want to increase sales they kind of know where they want to go but there's a lot of different choices. How is the analytics side of your product offering and solutions set addressing that hey I don't want to actually have to to use the good to great analogy where you get on the bus you have the right people but ultimately you might have to change roads to get there so data will be a key part of that versus a prefabricated here's the path and it may or may not be there so what is the key product and solutions that you guys have to help a customer saying okay I know where I want to do I might not know from the beginning on how to get there. Yeah I like your current analogy I actually use a diet analogy which is I can read all the diet books that I want but as I actually do something I change my actions it's gonna just be interesting information so there's a way that we're advancing our technology in analytics so most often people think about reporting dashboard analysis still very vital for the strategic management reporting that we do. Where the next step is really on their journey is from that Cognos business intelligence world to predictive so SPSS software technology that's gonna help look and identify patterns that human eyes may not see make correlations and then predict the future so that's the SPSS software portfolio hoping to answer what could happen. The very next step after that is to put that together with business rules and say what do I want to happen so what should be the next best action that's using SPSS technology, I log technology to bring that together and then all the way to Watson which we know is really the cognitive can I learn what's best from the past and can I take that learning to make my decisions more and more often so there's a whole portfolio of analytics technology and a journey that clients are going on to go from historical views to looking at the future and then actually determining how to change the outcome for the better. So Harry we've seen IOD evolve over the years and last year we made the statement that IBM is essentially super glued it's analytics business to the big data meme so you saw the Hadoop movement grow up you guys observed that and then brought your portfolio to bear. I wonder if you could talk about how that decision came about to sort of how you essentially have created a new category big data analytics now everybody's talking about it but they weren't a year ago even it's amazing how fast things change in this space but I wonder if you could take us back and talk about the decision that you guys made to actually approach the market that way and I'd like to talk about the marketplace a little bit too. Sure so the keynote this morning said it very nicely which is big data is like my muscles that don't go to the gym they need to be exercised and it's analytics that exercises it so you can have all the data in the world but if you don't have good analytics to make sense of it it's kind of useless you can have the best analytics sophisticated algorithms in the world but if you don't have data to fuel it you're not gonna get the right decision out of it so really we saw the marriage very early on you need to put big data and analytics together and you mentioned Hadoop that's a great example we have infosphere big insights a great technology four times faster than other Hadoop distributions out there great technology we've married that together with our products from the SPSS portfolio something like analytic catalyst people worry about they have to code things they have to have data scientists well analytic catalyst you pointed at big insights and it's gonna give you English language interpretation it's gonna let the data do the talking it's just a fantastic sort of marriage to put our big data technologies together with analytics so how do you guys look at the market the TAM you got all kinds of different players you got large companies like yourself you know Oracle's now sort of embracing the big data you got the sort of pure play elephants we sometimes call them I cite Cloudera and Hortonworks and others get some small players emerging what's IBM's perspective on the developments in the marketplace how do you look at it so we see ourselves very unique so yes there's definitely the sort of mega vendors out there which are looking to almost tie our customers into a particular stack so they're tying them to the Oracle stack the SAP stack what we want to do is we want to say look you got to take an outcomes driven approach it starts with a business problem you need analytics and you need big data and you need infrastructure that understands those workloads however we recognize clients already have an existing investment so we want to work with what clients already have we want to expand their technology footprint their IT foundation with it streams or with Hadoop or with better data warehousing to be better performance so start with what people have expand to address that business problem and grow from there so we believe we have a very different position to the other mega vendors who are really looking for everybody to shift to their technology one of the aspects of what you just mentioned is openness the word open used to be of it's changed over the years I had Lou Gerstner called you guys a recovering alcoholic when it came to things like locking customers in we were at a meeting last week we heard Joe Tucci one of your competitors at one of the big companies essentially say we're the open company those guys pointing at you and Oracle and others are the closed guys trying to get people into the stack so you're saying no that's not the case let's talk about more detail about that so you're basically approaching openness through open APIs open standards I wonder if you could talk about that a little bit because as a customer you get confused everybody's pointing fingers saying oh we're open they're closed and we're open they're closed what gives you confidence that you know you're open yep so we do a number of things in our technology to make sure we're open first of all we are large contributors to open source and open stacks so we have got investments around contributing to the open source community and we're very serious about our investments there secondly we have open APIs and open things like big sequel against Hadoop which everybody can use what they have invested so our openness is really around IT have got serious investments already in their IT shop we understand that we're not looking for them to replace what's working for them we're looking to be able to flesh out what they need in order to address a business problem because big data and analytics it's not going to be one technology it's not going to be one type of analytics that's going to solve a business problem it's the ability to use them all in combination that's key Harry we were just in New York City for our big data NYC event there was a lot of announcements that was pretty geeky as you talk about Hadoop and things like that but we're seeing the the early adopters crossover to not just production scaling say Hadoop or other big data and analytics to get that real time piece and then behind that is the followers they're going to come in the second wave which is massive adoption so one do you agree that you're seeing those early sets of adopters scaling into production because some of the use cases are showing their hand or not and two what is the use cases that you see right now for businesses that are the step one the kindergarten elementary school where they are going to have business value they're the low-hanging fruit for the use cases yeah so there's a lot of clients out there I would call it kicking the tires so there's a tremendous amount that you can download and try for yourself in terms of testing out your skill sets testing out Hadoop we have the ability to download our technology as well for people who are just kind of trying it out for themselves now where those projects get stuck is they're in that pilot phase really how do you put them into production and that's where the specific high-value use cases come in so we see a lot of customers focused in two main areas the first is around understanding the customer customer lifetime value customer retention understanding how to deliver a superior customer experience and that's where we can bring together the technology in order to address that particular domain the second area is in operation so the ability to sort of streamline and gain efficiency in operations so we have a signature solution particularly around our predictive maintenance so cars, factory lines you want to predict when they're going to fail and put the maintenance in right there before it fails because it's extremely costly to jump in and fix something once it's broken so predictive maintenance it's kind of a concept right that we've been hearing about the industrial internet or internet of things right the equipment, the machine to machine data yes and the great thing is if you're at the conference you can go down and see the connected car where the connected car will be telling you hey it's the person in front of you slowing down you should put on your brakes in fact it may put on the brakes for you automatically in the future it's going to be connected in that we are we can track hundreds of thousands of sensors about where that car is how that car is feeling as an internet of a thing and be able to bring it in for service be able to tell people how to run it efficiently and then insurance companies are using that data to say you know what's the insurance rate going to be for a driver of that nature if there a speeder may be or if there a dangerous driver we can adjust insurance rates for that example Harry what do you see the demand for these technologies these initiatives coming from within organizations is it the traditional IT department is it the business you know people put out the I think it was Gartner put out the data the CMO is going to spend more than CIOs right good fodder, good discussion points and there seems to be legitimately a lot of action outside of the IT organization what are you seeing so this is where we we see the big data analysis coming together to an IT person who's talking about big data they're normally talking about big data and the analytics that apply to it they're using big data almost as shorthand and a lot of IT people are looking at how can I use the new technologies in big data to really gain economies in IT as well as better treat the data how it wants to be treated which is if it's unstructured it likes to stay unstructured that's why Hadoop exists you don't need to model it if it's streaming I want to act on it when it's streaming if it's to be stored and modeled I want to warehouse to do that so IT are looking at the technologies in big data to flesh out the IT infrastructure but you're right the business side of the house may not be calling it a big data project they may actually be calling it an analytics project but they mean big data because they want more data in their analytics project or a new revenue project here exactly so what do you say to the CIO who may be seeing this movie before he or she is a skeptic they've made technology investments maybe they haven't panned out this is a big trend they read the Gartner hype cycle you know, et cetera, et cetera, et cetera and they say, you know what? I'm just going to sit back and wait it's a good idea, a bad idea why or why not so I don't know that waiting is a good approach the strategy is just not a very good one is that anything? me waiting for the next possible best iPhone I might as well just have a phone I can use right now so the place that we say to start is look for a part in your business that can get a dramatic return from investment and we look at three main areas first off, are there areas of my business that if I could provide more complete answers I'm going to be able to make better decisions so I may already be making good decisions today but can I fuel that with more insight with more data with better more sophisticated analytics can I get more complete answers the second area is where can I improve business processes by applying big data and analytics to that business process can I reduce my follow up on fraud by automating some of the fraud that isn't important and only having the people following up on fraud on the pieces that are going to provide good value so can I tune my resources and business processes to be better deployed the third area is actually whole new business model so I'll go back to your couponing example so a telco is always interested in how to retain customers and how to provide better services on those mobile devices we see now new business models appearing between telcos and retail to say well if I could partner up in a new business model between my mobile device and my retail store I can offer coupons knowing where that mobile device is knowing the person's walking past the store and offer them something to incent them to go into the store so we see whole new business models happening because of this ability because of the real time nature of what you can do today versus what you know I mean in the early days you remember the early days of decision support right some of these three not the third one not the new business models but certainly the more complete answers remember the 360 degree view of your business and to a certain extent business process improvements through analytics those were sort of promises of the old analytics business the new analytics business you know is promising even more new business model changes what gives you confidence that we'll actually see that vision this time through I have confidence in it because we're already seeing it today at customers in that the sort of traditional let's say traditional analytics was in that manager's hands right I'm going to monitor the business I'm going to see if I'm performing against my metrics for the business but we're now seeing analytics in the actual business processes themselves so we have a trucking company they're actually providing analytics to the drivers where's the best price for petrol where's the best location you should be stopping now for your rest stop so we're actually seeing people put analytics into the hands of people you would not have thought about even five years ago you would be able to enable those people and they're not thinking of themselves as information workers or I'm going to stop to make a decision now it's doing a job there's no longer a difference between analytics and operations there's analytics and there's analytical operations we have some questions coming up in our crowd chat here crowd cap is going to be proud of Tim Tim good job out there is participating he says if a company could improve business process or value why wouldn't they have been done done it before has tech slash big data really been the hurdle so there's a couple of reasons why people haven't done it before I said I love that question it wasn't possible before one databases right one was the data may not have been captured before right we were all chit chatting before and talking on our mobile devices before but now we can capture everything that's happening on Twitter and Facebook so sometimes the data just wasn't available secondly it was cost prohibitive to actually take advantage of that data so new infrastructure new database storage new file system storage and made it more cost effective to actually pursue that data and make sense of it and third the technology in analytics has advanced I mean I don't know about you but I'm not a statistician that can identify correlations and sentiment and text I need technology to do that so the technology has advanced to do it too so I think it's just that we're at the cusp of things are now possible that weren't possible before that's why everyone talks about that we're in early days or bottom of the first inning whatever analogy people want to use really it's all three right the instrumentation capabilities just weren't always there in real time and doctor Tim Buckman was saying that one of the most important things that they look at is the accuracy of the data and he made metaphors to the airline industry so those three things kind of have been the perfect storm so I gotta ask you in respect to the technology you have new cases evolving in real time so there's some low hanging fruit we talked about that but the technology theater things are happening really quickly so how do you as a company I'd be imbalanced that even some stack discussions going on on crowd chat so is it the right stack open close the iPhone of the data center open Android iPhone analogies so you got technology changing very very fast and use cases are evolving where some say the risk is that you could be in a cul-de-sac we're no customers so you know it's a challenge how do you guys balance that so yes so every IT person I think is here who worries about agility and the amount of expenditure they have to invest in new projects versus maintaining or keeping the shop up and running so we care a lot about agility and one way to have agility is to and we advise clients to do this which is invest in a big data analytics platform but do it one project at a time you're not gonna solve remember the whole SOA SOA of 10 years ago you know I'm gonna re-architect everything from the ground up that's just not reality today what we want to do is we want to say have a roadmap or a mental map of where you're heading and the infrastructure and foundation you need for that it's not that you're gonna implement it all day one it's that you want to you want to take each project on a roadmap by design not just let it happen to you and that's gonna give you agility because you're not just building for one project I'm building for my customer center now I'm building for my operations now I'm building for finance you're building a foundation with your first project that you can just expand from project to project and that's where you get the return so talk about the shadow IT market because Dave and I we've been talking about this for many many years now shadow IT this is the term where IT goes around in their own processes to go play in the cloud and what's been interesting is that it's become almost legit and kind of don't ask don't tell policy where you can get stuff done you can do things in there and be agile where there's not a lot of reconstruction going on and so people are seeing those use cases of hey I can go into the cloud and if it breaks I don't have collateral damage with infrastructure so that notion of being agile not just test and dev moving apps into the crowd as so comments I want if you see that and two where does the integration because the conversation around integrated stacks always comes in at what point does the application have what components of the data fabric so shadow IT is an experiment for agile has that going to impact IT will it become legitimate and two where does the analytics sit in the application okay so that's a quite a big topic for a few minutes so in terms of IT I actually believe we're at the cusp of redefining the definition IT and IT I feel is very much right now defined as I'm responsible for what's inside my organization I'm responsible for keeping those systems up and running delivering applications myself through that test of production if we look at IT as redefining themselves as saying I'm here as a business partner to deliver the systems the applications the data to my business regardless of whether it's in my house it's on the cloud or it's actually in partnership with someone else and their IT infrastructure then I think that we don't no longer call it shadow IT we call it a definition that IT's responsibility is to empower the business and they're doing it from a combination of internal systems systems on the cloud and partner systems so that's your first question which is I don't see it as shadow IT I see it as an expansion of the definition of IT so it's becoming legitimized in different ways not like go do things legitimate with credit cards but like as part of the business process now in terms of discretionary spend or shadow systems popping up because technology is very easy for a business person to maybe source them themselves and not go via IT for the requirements setting that is actually providing IT some of the agility that they're sort of lacking or still trying to catch up on which is if there's a cloud application that can satisfy a business need it shouldn't be seen potentially as a threat to IT we should be embracing it as part of the portfolio of the business demands otherwise they wouldn't be really looking for it where it comes down and what you point out is we need to be able to integrate those applications in with the IT systems in the shop to take advantage of them so there's a lot of work that's happening right now to say there's Twitter data, there's cloud data there's data sitting in apps that the business side of the house have acquired or buying a software as a service we need to bring that data into the fold as I said step by step and then deliver it as a whole so people always talk about data as a one-way street it was from systems to people and then people do something magical with it we're really seeing now it's a two-way street the data that's happening in the business the systems happening on the cloud they need to come back into the fold and be integrated and our technology and our information integration will help make that happen we're seeing that same trend where essentially like the customers see the destination point A to point B but a lot of times their question that they really need to be asked and comes after a few tries the real question is okay we now know after it's like Google search almost it's like not like a pre-canned queries in the old days remember the old days so we're seeing companies like Splunk and other companies where hey, here's some machine data we say trash to gold which is log files with that now as insight so what you guys have is the same concept where you can bring data in and integrate it up into the applications what you guys, is that a good explanation? Yes, yes, I think that the challenge that companies have is that right now there's this gold nugget in the data but we don't know what is a byproduct of the business in terms of data or what's a co-product in that I can leverage some of that data value so people have become hoarders of data, right? It's almost like a competition how much data can I call? That's a reality show, we should get on that with our data orders data orders but really the piece that's lacking is I don't see it as a I have too much data problem I see it as I don't have a good enough filter for that data I don't have enough filters in order to be able to understand what's valuable in it and what insight I can gain from it My final question for you is much more of a personnel question as we talk about data exhaust to data gold and that's some of the real innovators have done that have taken what looks like data exhaust or pollution if you want to look at it that way and turn it into real value or gold so the final question is what are you seeing the innovators out there? What's the makeup of the folks that are the ones that are seeing the early opportunities around taking data, not just hoarding it but like turning it into value? Do you see a new kind of executive a new kind of personnel category where there's a data scientist? I mean, what are some of the share with the folks some of the insight around the kind of personnel the person that's making things happen? Well, I think some of the people who are actually making it happen may not like this analogy but I'd say the people that are really making it happen thinking like six year olds and let me explain why so they're thinking like six year olds because if you have a small child they get to that age where every question is why why, why, why, why this, why that, why that and quite often they eventually get told because and really the people who are innovating in this market with big data and analytics are the people that are constantly questioning so they're like a six year old they constantly question and they're not just gonna say it's because because we always did it that way because that's the way the industry runs because that's the way that we work with inside this organization that people are really successful they break free from that and they're gonna be constantly questioning the status quo, constantly saying I don't know what I don't know and they're gonna use big data and analytics to get to that answer. That's great, Harry, great. Well, end on that note. We're in a toddler industry right now we're asking why. If we can have a quiz show ask who's smarter than a fifth grader we'll get Watson on here to compete but big data is changing the world from exhaust to value we're seeing it all over this is really about a business value with analytics on one hand and social business on the other with all the technology underneath great sound bites on crowdchat.net Harry thanks for joining us we'll be right back here in theCUBE but live at information on demand IBM's conference hashtag IBM IOD we'll be right back.