>> Live from Boston, it's theCUBE. Covering IBM Chief Data Officer Summit. Brought to you by IBM. >> Welcome back everyone to theCUBE's live coverage of the IBM Chief Data Summit here in Boston, Massachusetts. I'm your host, Rebecca Knight along with my co-host Paul Gillin. We're joined by Ash Dhupar, he is the Chief Analytics Officer at Publishers Clearing House. Thank you so much for coming on theCUBE. >> Thank you Rebecca for calling me here. >> So Publishers Clearing House is a billion-dollar company. We think of it as the sweepstakes company, we think of the giant checks and be the Prize Patrol surprising contestants, but it's a whole lot more than that. Tell our viewers a little bit, just explain all the vast amount of businesses that you're in. >> Sure, so, in a nutshell, we are a media and entertainment company with a large base of customers, about 100 million customers who are motivated with the chance to win. That's the sweepstakes angle to it. And we have, you can categorize the business into two buckets. One is our media and entertainment side, which is the publishing side. And then the other is our retail side which is where we sell merchandise to our customers. Think of us as a catalog and an e-commerce company. On the media and entertainment side, we have a very good engagement with our customers, we get about two billion page views on a monthly basis on our website. We, about 15 million unique customers on a monthly basis are coming to the site and they spend a considerable amount of time with us on an average, anywhere between 12 to 15 minutes, depending on, you know the type of the customers. Some of our very heavily-engaged customers can spend as much as about two hours a day with us. (Rebecca and John laughs) >> Trying to win that, that either the big prize or there are small prizes like, if you go on our site, there's a winner everyday, like there could be 1,000 dollar winner everyday playing a certain type of a game. So that's the media and the entertainment side of our business, that's completely ad-supported. And then we are the retail side of the business is we are in direct mail, so the traditional, we would send someone a direct mail package. And an e-commerce company as well. Just as a small nugget of information, we are. We send almost about 400 million pieces of physical mail which is including our packages that are sent and so on and so forth and though also still a large direct mail company. Still profitable and still growing. >> I'm sure the US Postal Service is grateful for your support. (laughs) They need all the help they can get. You collect, essentially, the prize money, is your cost of data acquisition and you have a huge database you told us earlier before we started filming of about 100 million people, that you have data on just in the US alone. Now what are you doing at the upper limits of what you're able to do with this data. How are you using this strategically other than just you know personalized email? >> Sure, so I think using data is a core asset for us. We are utilizing in giving our customers better experiences by utilizing the data we have on them. Marrying it with other data sources as well. So that we can personalize the experience. So that we can make your experience when you come on the site better. Or if we are sending something to you in mail, we give you products that are relevant to you. So to bring it down to a little more tactical level, in case of when you are on our site, then on our e-commerce site, there's a product recommendation engine, right? Which goes in and recommends products to you on what products to buy. Those product recommendation engines drive a significant amount of sales, almost about 40% of our sales are driven by the prior recommendation engines that is all understanding of the customer, what you're buying, what you're likely to buy and the algorithms behind it are built with that. >> Can you give another example though, of how, if I were, I mean you said all these customers are united by a common desire to win and to play a game and to win. >> Right. >> But what are some other ways beyond product recommendation engines, which are now sort of old hat. >> Right. >> What other ways are you enhancing the customers experience and personalizing it? >> Sure, sure. So, I'll give you a recent example of where we are utilizing some of the data to give a more relevant experience to the customer. So when a customer comes on our website, right when you're coming to register with us. So, as you register, as you fill in the form, after you give your name, address and your email address and you hit submit, at that very second, there are some algorithms that are running behind the scenes to understand how are you likely to engage with us. How are you going to, let's say, because we have a diverse business, are you likely to buy something from us? Or are you not likely to buy something from us? And if you're not likely to buy something from us, which means I can get you to, and you know not waste your time in showing you merchandise, but I can give you an experience of free-to-play games and you can, within free-to-play games, what type of games like understanding the persona of the person. We could say, hey, you probably are a lotto player or you are a word game puzzle player and we could give you and direct you to those experiences that are more relevant to you. In case of, if you're going to buy something from us, are you likely to buy, you know highly likely to buy or less likely to buy. Depending on that, should I show you just 10 or 15 products or should I show you like more than that? Are you more likely to buy a magazine? So making it more relevant for the customer experience is where it is all about. We use a lot of this data to, to make that happen. >> So analytics is really core to your business. It's the, completely strategic. Where do you sit in the organization, organizational layout, how is that reflected in the way your job is integrated into the organization? >> Sure, so, it is, I'm part of the C-Suite. And I think our CEO, he had this vision, thing he started. He loves data first of all. (laughs) >> Lucky for you. (laughs) >> Thank you. And he truly believes that data and analytics can drive growth and bring innovation from different areas if we utilize it in the best possible way. So A, I am part of that team. And work very closely with each of the business owners. That's the key, out here is like you know, it is, analytics is not in one corner but in the center of all the, all the business areas giving them either insights or building algorithms for them so that we can make either better decisions or we can power growth, depending on which way we are looking at it. >> You're the Chief Analytics Officer and we're here at the Chief Data Summit here, of here. How different are the roles in your mind and do they work together? I mean you have a CTO that is responsible for sort of Chief Data Officer. >> Yes. >> Responsibilities. How do you two collaborate and work together? >> It is a very tight collaboration. And they're two separate jobs but it is a very tight collaboration, we work hand in hand with each other. And the best part I would say is that you know, we're all focused and we're all driving towards how can we drive growth? That's the bottom line, that is where the bucks stops for all of us in the companies. Are we building projects? Are we doing things that is going to grow the company or not? So the collaboration with the CTO is A, a critical piece. They own the infrastructure, as well as the data and when you own the data, which is, in a way, is slightly, I would say, data governance I would say is a thankless job (laughs) believe it or not. But it is a critical job. It is if your data is not right, it is not going to work for whatever you're trying to do, it's the garbage in garbage out, we all know about that. And we work very closely. If there are CAPEX proposals that needs to be put in place because we're going after a certain big project, whether it's putting things together in one place or a 360 view of the customer. All of that is worked hand in hand. We work together in working towards that. >> What is your big data infrastructure like? Is it on the Cloud? Is it your own? Are you Adobe based? What do you use? >> All of the above. >> Oh. (laughter) No, so, what we have is because we are such an old company, you know we still have our legacy Db2 infrastructure. A lot of our backend databases, lot of our backend processes are all attached to that. We have a warehouse, a sequel server warehouse. We also, for our web analytics, we use Google's BigQuery. That's where you collect a lot of data on a daily basis. And recently, I think about three years ago, we went into the Cloud environment. We have a map, our cluster, which was cloud-based and now, we have brought in on prem very recently. >> Back from the Cloud. >> Back from the Cloud, on prem. And there was very good reasoning why we did that. I think frankly, it's cheaper on a longer term to bring that on prem and you are a lot more in control with all the issues with data privacy. So it is. >> Which, I hope you don't mind my interrupting but we have to wrap here and I need to get that question in. (laughs) >> Yes. >> You have data on 100 million consumers. What are you doing with all of the attention being paid for privacy right now? What are you doing to ensure the. >> We have a very, very I would say integrated infrastructure, data governance, data. There's a whole slew of, I would say, people and process around that to make sure that our date is not exposed. Now luckily, it's it's not like PII to the level that it's a health care data. So you are not really, you have information that is crazy but you still have the PII, the name and address of these customers. And as an example, none of the PII data is actually available to even to the analytics folks. It's all stripped, the PII's stripped off. You give us an ID to the customer and frankly the analytics team don't need the PII information to build any algorithms as well. So there is a whole process around keeping the data secure. >> Great, well Ash, thank you so much for coming on theCUBE, it was a pleasure having you. >> Thank you and thank you for inviting me. >> I'm Rebecca Knight for Paul Gillin. We will have more from IBM CDO Summit just after this. (techno music)

>> Narrator: Live, from San Jose, it's theCUBE. Presenting Big Data Silicon Valley. Brought to you by SiliconANGLE Media and its ecosystem partners (mellow electronic music) >> Welcome back to theCUBE, on day two of our coverage of our event, Big Data SV. I'm Lisa Martin, my co-host is Peter Burris. We are the down the street from the Strata Data Conference, we've had a great day yesterday, and great morning already, really learning and peeling back the layers of big data, challenges, opportunities, next generation, we're welcoming back to theCUBE an alumni, the CMO of Kinetica, Dan Raskin. Hey Dan, welcome back to theCUBE. >> Thank you, thank you for having me. >> So, I'm a messaging girl, look at your website, the insight engine for the extreme data economy. Tell us about the extreme data economy, and what is that, what does it mean for your customers? >> Yeah, so it's a great question, and, from our perspective, we sit, we're here at Strata, and you see all the different vendors kind of talking about what's going on, and there's a little bit of word spaghetti out there that makes it really hard for customers to think about how big data is affecting them today, right? And so, what we're actually looking at is the idea of, the world's changed. That, big data from five years ago, doesn't necessarily address all the use cases today. If you think about what customers are going through, you have more users, devices, and things coming on, there's more data coming back than ever before, and it's not just about creating the data driven business, and building these massive data lakes that turn into data swamps, it's really about how do you create the data-powered business. So when we're using that term, we're really trying to call out that the world's changed, that, in order for businesses to compete in this new world, they have to think about to take data and create CoreIP that differentiates, how do I use it to affect the omnichannel, how do I use it to deal with new things in the realm of banking and Fintech, how do I use it to protect myself against disruption in telco, and so, the extreme data economy is really this idea that you have business in motion, more things coming online ever before, how do I create a data strategy, where data is infused in my business, and creates CoreIP that helps me maintain category leadership or grow. >> So as you think about that challenge, there's a number of technologies that come into play. Not least of which is the industry, while it's always to a degree been driven by what hardware can do, that's moderated a bit over time, but today, in many respects, a lot of what is possible is made possible, by what hardware can do, and what hardware's going to be able to do. We've been using similar AI algorithms for a long time. But we didn't have the power to use them! We had access to data, but we didn't have the power to acquire and bring it in. So how is the relationship between your software, and your platform, and some of the new hardware that's becoming available, starting to play out in a way of creating value for customers? >> Right, so, if you think about this in terms of this extreme data concept, and you think about it in terms of a couple of things, one, streaming data, just massive amounts of streaming data coming in. Billions of rows that people want to take and translate into value. >> And that data coming from-- >> It's coming from users, devices, things, interacting with all the different assets, more edge devices that are coming online, and the Wild West essentially. You look at the world of IoT and it's absolutely insane, with the number of protocols, and device data that's coming back to a company, and then you think about how do you actually translate this into real-time insight. Not near real-time, where it's taking seconds, but true millisecond response times where you can infuse this into your business, and one of our whole premises about Kinetica is the idea of this massive parallel compute. So the idea of not using CPUs anymore, to actually drive the powering behind your intelligence, but leveraging GPUs, and if you think about this, a CPU has 64 cores, 64 parallel things that you can do at a time, a GPU can have up to 6,000 cores, 6,000 parallel things, so it's kind of like lizard brain verse modern brain. How do you actually create this next generation brain that has all these neural networks, for processing the data, in a way that you couldn't. And then on top of that, you're using not just the technology of GPUs, you're trying to operationalize it. So how do you actually bring the data scientist, the BI folks, the business folks all together to actually create a unified operational process, and the underlying piece is the Kinetica engine and the GPU used to do this, but the power is really in the use cases of what you can do with it, and how you actually affect different industries. >> So can you elaborate a little bit more on the use cases, in this kind of game changing environment? >> Yeah, so there's a couple of common use cases that we're seeing, one that affects every enterprise is the idea of breaking down silos of business units, and creating the customer 360 view. How do I actually take all these disparate data feeds, bring them into an engine where I can visualize concepts about my customer and the environment that they're living in, and provide more insight? So if you think about things like Whole Foods and Amazon merging together, you now have this power of, how do I actually bridge the digital and physical world to create a better omnichannel experience for the user, how do I think about things in terms of what preferences they have, personalization, how to actually pair that with sensor data to affect how they actually navigate in a Whole Foods store more efficiently, and that's affecting every industry, you could take that to banking as well and think about the banking omminchannel, and ATMs, and the digital bank, and all these Fintech upstarts that are working to disrupt them. A great example for us is the United States Postal Service, where we're actually looking at all the data, the environmental data, around the US Postal Service, we're able to visualize it in real-time, we're able to affect the logistics of how they actually navigate through their routes, we're able to look things like postal workers separating out of their zones, and potentially kicking off alerts around that, so effectively making the business more efficient. But, we've moved into this world where we always used to talk about brick and mortar going to cloud, we're now in this world where the true value is how you bridge the digital and physical world, and create more transformative experiences, and that's what we want to do with data. So it could be logistics, it could be omnichannel, it could be security, you name it. It affects every single industry that we're talking about. >> So I got two questions, what is Kinetica's contribution to that, and then, very importantly, as a CMO, how are you thinking about making sure that the value that people are creating, or can create with Kinetica, gets more broadly diffused into an ecosystem. >> Yeah, so the power that we're bringing is the idea of how to operationalize this in a way where again, you're using your data to create value, so, having a single engine where you're collecting all of this data, massive volumes of data, terabytes upon terabytes of data, enabling it where you can query the data, with millisecond response times, and visualize it, with millisecond response times, run machine learning algorithms against it to augment it, you still have that human ability to look at massive sets of data, and do ad hoc discovery, but can run machining learning algorithms against that and complement it with machine learning. And then the operational piece of bringing the data scientists into the same platform that the business is using, so you don't have data recency issues, is a really powerful mix. The other piece I would just add is the whole piece around data discovery, you can't really call it big data if, in order to analyze the data, you have to downsize and downsample to look at a subset of data. It's all about looking at the entire set. So that's where we really bring value. >> So, to summarize very quickly, you are providing a platform that can run very, very fast, in a parallel system, and memories in these parallel systems, so that large amounts of data can be acted upon. >> That's right. >> Now, so, the next question is, there's not going to be a billion people that are going to use your tool to do things, how are you going to work with an ecosystem and partners to get the value that you're able to create with this data, out into the engine enterprise. >> It's a great question, and probably the biggest challenge that I have, which is, how do you get above the word spaghetti, and just get into education around this. And so I think the key is getting into examples, of how it's affecting the industry. So don't talk about the technology, and streaming from Kafka into a GPU-powered engine, talk about the impact to the business in terms of what it brings in terms of the omnichannel. You look at something like Japan in the 2020 Olympics, and you think about that in terms of telco, and how are the mobile providers going to be able to take all the data of what people are doing, and to related that to ad-tech, to relate that to customer insight, to relate that to new business models of how they could sell the data, that's the world of education we have to focus on, is talk about the transformative value it brings from the customer perspective, the outside-in as opposed to the inside-out. >> On that educational perspective, as a CMO, I'm sure you meet with a lot of customers, do you find that you might be in this role of trying to help bridge the gaps between different roles in an organization, where there's data silos, and there's probably still some territorial culture going on? What are you finding in terms of Kinetica's ability to really help educate and maybe bring more stakeholders, not just to the table, but kind of build a foundation of collaboration? >> Yeah, it's a really interesting question because I think it means, not just for Kinetica, but all vendors in the space, have to get out of their comfort zone, and just stop talking speeds and feeds and scale, and in fact, when we were looking at how to tell our story, we did an analysis of where most companies were talking, and they were focusing a lot more on the technical aspirations that developers sell, which is important, you still need to court the developer, you have community products that they can download, and kick the tires with, but we need to extend our dialogue, get out of our customer comfort zone, and start talking more to CIOs, CTOs, CDOs, and that's just reaching out to different avenues of communication, different ways of engaging. And so, I think that's kind of a core piece that I'm taking away from Strata, is we do a wonderful job of speaking to developers, we all need to get out of our comfort zone and talk to a broader set of folks, so business folks. >> Right, 'cause that opens up so many new potential products, new revenue streams, on the marketing side being able to really target your customer base audience, with relevant, timely offers, to be able to be more connected. >> Yeah, the worst scenario is talking to an enterprise around the wonders of a technology that they're super excited about, but they don't know the use case that they're trying to solve, start with the use case they're trying to solve, start with thinking about how this could affect their position in the market, and work on that, in partnership. We have to do that in collaboration with the customers. We can't just do that alone, it's about building a partnership and learning together around how you use data in a different way. >> So as you imagine, the investments that Kinetica is going to make over the next few years, with partners, with customers, what do you hope Kinetica will be in 2020? >> So, we want it to be that transformative engine for enterprises, we think we are delivering something that's quite unique in the world, and, you want to see this on a global basis, affecting our customer's value. I almost want to take us out of the story, and if I'm successful, you're going to hear wonderful enterprise companies across telco, banking, and other areas just telling their story, and we happen to be the engine behind it. >> So you're an ingredient in their success. >> Yes, a core ingredient in their success. >> So if we think about over the course of the next technology, set of technology waves, are they any particular applications that you think you're going to be stronger in? So I'll give you an example, do you envision that Kinetica can have a major play in how automation happens inside infrastructure, or how developers start seeing patterns in data, imagine how those assets get created. Where are some of the kind of practical, but not really, or rarely talked about applications that you might find yourselves becoming more of an ingredient because they themselves become ingredients to some of these other big use cases? >> There are a lot of commonalities that we're starting to see, and the interesting piece is the architecture that you implement tends to be the same, but the context of how you talk about it, and the impact it has tends to be different, so, I already mentioned the customer 360 view? First and foremost, break down silos across your organization, figure out how do you get your data into one place where you can run queries against it, you can visualize it, you can do machine learning analysis, that's a foundational element, and, I have a company in Asia called Lippo that is doing that in their space, where all of the sudden they're starting to glean things they didn't know about their customer before to create, doing that ad hoc discovery, so that's one area. The other piece is this use case of how do you actually operationalize data scientists, and machine learning, into your core business? So, that's another area that we focus on. There are simple entry points, things like Tableau Acceleration, where you put us underneath the existing BI infrastructure, and all of the sudden, you're a hundred times faster, and now your business folks can sit at the table, and make real-time business decisions, where in the past, if they clicked on certain things, they'd have to wait to get those results. Geospatial visualization's a no-brainer, the idea of taking environmental data, pairing it with your customer data, for example, and now learning about interactions. And I'd say the other piece is more innovation driven, where we would love sit down with different innovation groups in different verticals and talk with them about, how are you looking to monetize your data in the future, what are the new business models, how does things like voice interaction affect your data strategy, what are the different ways you want to engage with your data, so there's a lot of different realms we can go to. >> One of the things you said as we wrap up here, that I couldn't agree with more, is, the best value articulation I think a brand can have, period, is through the voice of their customer. And being able to be, and I think that's one of the things that Paul said yesterday is, defining Kinetica's success based on the success of your customers across industry, and I think really doesn't get more objective than a customer who has, not just from a developer perspective, maybe improved productivity, or workforce productivity, but actually moved the business forward, to a point where you're maybe bridging the gaps between the digital and physical, and actually enabling that business to be more profitable, open up new revenue streams because this foundation of collaboration has been established. >> I think that's a great way to think about it-- >> Which is good, 'cause he's your CEO. >> (laughs) Yes, that sustains my job. But the other piece is, I almost get embarrassed talking about Kinetica, I don't want to be the car salesman, or the vacuum salesman, that sprinkles dirt on the floor and then vacuums it up, I'd rather us kind of fade to the behind the scenes power where our customers are out there telling wonderful stories that have an impact on how people live in this world. To me, that's the best marketing you can do, is real stories, real value. >> Couldn't agree more. Well Dan, thanks so much for stopping by, sharing what things that Kinetica is doing, some of the things you're hearing, and how you're working to really build this foundation of collaboration and enablement within your customers across industries. We look forward to hearing the kind of cool stuff that happens with Kinetica, throughout the rest of the year, and again, thanks for stopping by and sharing your insights. >> Thank you for having me. >> I want to thank you for watching theCUBE, I'm Lisa Martin with my co-host Peter Burris, we are at Big Data SV, our second day of coverage, at a cool place called the Forager Tasting Room, in downtown San Jose, stop by, check us out, and have a chance to talk with some of our amazing analysts on all things big data. Stick around though, we'll be right back with our next guest after a short break. (mellow electronic music)

okay we're back live here inside the cube rounding out day one of exclusive coverage of IBM information on demand I'm John further the founder SiliconANGLE enjoy my co-host Davey lonte we're here in heat you saw who's the vice president I said that speaks that you know I think you always get promoted you've been on the cube so many times you doing so well it's all your reason tatian was so amazing I always liked SVP the cute good things happen that's exactly why i be MVP is a big deal unlike some of the starters where everyone gets EVP all these other titles but welcome back thank you so the storytelling has been phenomenal here although murs a little bit critical some of the presentations earlier from gardner but the stories higher your IBM just from last year take us through what's changed from iod last year to this year the story has gotten tighter yes comprehensive give us the quick okay quick view um okay here's the point of view here's the point of view first you got to invest in a platform which we've all talked about and i will tell you it's not just us saying it i would say other vendors are now copying what we're saying cuz if you went to strata yes which you were there we were there probably heard some of the messages that's right why everybody wants to be a platform okay one two elevated risk uncertainty governance I think privacy privacy security risk this is what people are talking about they want to invest in a more why because you know what the decisions matter they want to make bigger beds they want to do more things around customer experience they want to improve products they want to improve pricing the third area is really a cultural statement like applying analytics in the organization because the people and the skills I would say the culture conversation is happening a lot more this year than it was a year ago not just at IOD but in the industry so I think what you're seeing here at IOD is actually a reflection of what the conversations are happening so our organizations culturally ready for this I mean you guys are going to say yes and everybody comes on says oh yes we're seeing it all over the place but are they really ready it depends I think some are some are absolutely ready some are not and probably the best examples are and it really depends on the industry so I'll give you a few examples so in the government area I think people see the power of applying things like real-time contextual insight leveraging stream computing why because national security matters a lot of fraudulent activity because that's measurable you can drive revenue or savings healthcare people know that a lot of decision-making is being made without a comprehensive view of the analytics and the data now the other area that's interesting is most people like to talk about text analytics unstructured data a lot of social media data but the bulk of the data that's actually being used currently in terms of big data analytics is really transactional data why because that's what's maintained in most operational systems where health systems so you're going to see a lot more data warehouse augmentation use cases leverage you can do on the front end or the back end you're going to see kind of more in terms of comprehensive view of the customer right augmenting like an existing customer loyalty or segmentation data with additional let's say activity data that they're interacting with and that was the usta kind of demo showing social data cell phone metadata is that considered transactional you know it is well call me to record right CDR call detail records well the real time is important to you mentioned the US open just for folks out there was a demo on stage when you guys open data yeah at all the trend sentiment data the social data but that's people's thoughts right so you can see what people are doing now that's big yeah you know what's amazing about that just one second which is what we were doing was we were predicting it based on the past but then we were modifying it based on real time activity and conversation so let's say something hot happened and all of a sudden it was interesting when Brian told me this he was like oh yeah Serena's average Twitter score was like 2,200 twit tweets a day and then if some activity were to happen let's say I don't know she didn't he wrote she had got into a romance or let's say she decided to launch a new product then all of a sudden you'd see an accused spike rate in activity social activity that would then predict how they wanted to operate that environment that's amazing and you know we you know we love daily seen our our crowd spots be finder we have the new crowd chat one and this idea of connecting consumers is loose data it's ephemeral data it's transient data but it's now capture will so people can have a have fun into tennis tournament and then it's over they go back home to work you still have that metadata we do that's very kind of its transient and ephemeral that's value so you know Merv was saying also that your groups doing a lot of value creation let's talk about that for a second business outcomes what do you what's the top conversation when you walk into a customer that says hey you know here's point a point B B's my outcome mm-hmm one of those conversations like I mean what are they what are some of the outcomes you just talked to use case you tell customers but like what did some of the exact you know what I'll tell you one use case so and this was actually in the healthcare hotel you won healthcare use case in one financial services use case both conversations happened actually in the last two weeks so in the healthcare use case there's already let's say a model that's happening for this particular hospital now they have a workflow process typically in a workflow process you you're applying capabilities where you've modeled out your steps right you do a before be before see and you automate this leveraging BPM type capabilities in a data context you don't actually start necessarily with knowing what the workflow is you kind of let the data determine what the workflow should be so in the this was in an ICU arena historically if you wanted to decide who was the healthiest of the patients in the ICU because you had another trauma coming in there was a workflow that said you had to go check the nurses the patient's profile and say who gets kicked out of what bed or moved because they're most likely to be in a healthy state that's a predefined workflow but if you're applying streams for example all the sudden you could have real-time visibility without necessarily a nurse calling a doctor who that calls the local staff who then calls the cleaning crew rate you could actually have a dashboard that says with eighty percent confidence beds2 and ate those patients because of the following conditions could be the ones that you are proactive in and saying oh you know what not only can they be released but we have this degree of confidence around them being because of the days that it's coming obvious information that changes then potentially you know the way your kind of setting your rules and policies around your workflow another example which was really a government use case was think about in government security so in security scenarios and national security state there is you never quite know exactly what people are intended to do other than you know they're intending something bad right and they're intentionally trying not to be found so human trafficking it's an ugly topic but I want to bring it up for a second here what you're doing is you're actually looking at data compositions and and different patterns and resolving entities and based on that that will dictate kind of potentially a whole new flow or a treatment or remediation or activity or savior which is not the predefined workflow it's you're letting the data actually all of a sudden connect to other data points that then you're arriving at the insight to take the action where is completely different I wanna go back to sleep RFI course not healthcare examples yeah so where are we today is that something that's actually being implemented is that something they sort of a proof of concept well that's actually being done at it's being done in a couple different hospitals one of which is actually in hospital in Canada and then we're also leveraging streams in the emory university intensive Timothy Buckman on you did earlier oh yeah the ICU of the future right absolutely brilliant trafficking example brings up you know Ashley that's the underbelly of the world in society but like data condition to Jeff Jonas been on the queue as you know many times and he talks with his puzzle pieces in a way that the data is traveling on a network a network that's distributed essentially that's network computing I mean estate management so look at network management you can look at patterns right so so that's an interesting example so that begs the next question what is the craziest most interesting use case you seen oh my gosh okay now i got i think about oh yes and you can talk about and i can talk about that creates business value or society value oh you know I okay um for you are putting me on the spot the craziest one so 3 we could be great could be g-rated don't you know they go to 2k yeah you know what I participated three weeks ago tiaa-cref actually hosted a fraud summit where it was all investigators like they were doing crime investigation so more than sixty percent of the guys in the room carried weapons because they were Security Intelligence they were pleased they were DA's they repented I was not packing anyway and there was about so 60-plus percent were those right and then only about thirty percent in the room were what i would consider the data scientists in the room like these are the guys are trying to decide which claims are not true or false so forth there were at least like three or four use cases in that discussion that came out they were unbelievable so one is in the fraud area in particular and in crime they're luring the data there what does luring the data they're taking location-based data for geographic region they're putting crime data on top of that right historical like drug rings and even like datasets in miami-dade county the DA told me they were doing things where rather than looking at people that are doing the drugs they they realize people that had possession of a drug typically purchased within a certain location and they had these abandoned properties and were able to identify entire rings based on that another one this is also semi drug-related is in the energy utility space there was in the middle part of the United States houses in Nice urban areas where they were completely torn apart on the interior and build into marijuana houses and so of course they're utilizing high levels of gas and electricity in order to maintain the water fertilization everything else well what happens is it drives peaks in the way that the energy utility looks on a given day pattern so based on that they're able to detect how inappropriate activities are happening and whether it's a single opportunistic type activity whether it's saying this was doing laundry or irrigating the Erie hey we well you know what's interesting about electricity to is especially someone's using electricity but no one's like using any of the gas you're like home but no one's cooking you know something's a little long but it was fascinating i mean really fascinating there were like several other crime scenarios in terms of speed i actually did not know the US Postal Service is like the longest running federal institution that actually tracked like mail fraud and one of the use cases i'm sure jeff has talked about here on the cube is probably a moneygram use case but we talked about that we talked I mean it the stories were unreal because I was spending time with forensic scientists as well as forensic investigators and that's a completely do we're getting we're getting the few minutes need for a platform to handle all this diversity so that's the security risk the governance everything you gotta go cuz your star for the analyst me I can't watch this conversation one final question one of the best yet as we get drugs in there we got other things packing guns guns and drugs you in traffic you know tobacco if you go / news / tobacco well write the knowledge worker all right final question for I know you gotta go this big data applications were you know the guys in the mailroom the guys work for the post office are now unable to actually do this kind of high-level kind of date basically data science yeah if you will or being an analyst so that what I want you to share the folks your vision of the definition of the knowledge worker overused word that's been kicked around for the PC generates but now with handheld with analytical real-time with streaming all this stuff happening at the edge how is it going to change that the knowledge work or the person in the trenches it could be person the cubicle the person on the go the mobile sales person or anyone you know I some people feel threatened when they hear that you're going to apply data and analytics everywhere because you're it implies that you're automating things but that's actually not the value the real value is the insight so that you can double down on the decisions you want to make so if you're more confident you're going to take bigger bets right and decision-making historically has been I think reserved for a very elite few and what we're talking about now is a democratization of that insight and with that comes a lot of empowerment a lot empowerment for everyone and you don't have to be a data scientist be able to be able to make decisions and inform decisions if anything you know actually Tim Buckman I had a good conversation about them as a professional you know what I if I was a physician I'd want to work at the hospital that has the advanced capabilities why because it allows me as a professional physician to then be able to do what I was trained to do not to detect and have to pay attention to all these alarms going off you know I want to work at the institutions and organizations that are investing appropriately because it pushes the caliber of the work I get to do so I think it just changes the dynamics for everyone tim was like a high-priced logistics manager you want to work with people want to work with leaders and now we're in a modern era this new wave is upon us who care and they want to improve and this is about continuing to improve Dave and I always talk about the open source world that those principles are going mainstream to every aspect of business collaboration openness transparency not controlled absolutely absolutely Indy thanks so much for coming in the queue and know you're busy think of your time we are here live in the cube getting all the signal from the noise and some good commentary at the end a one we have one more guest ray way right up next stay tuned right back the queue