Big Data Analytics: The Revolution Has Just Begun



good morning I have the privilege of introducing our next keynote speaker he's the president and CEO of Lync analytics and also the co-founder now if you didn't know Lync analytics before you walked in this morning I guarantee you're going to remember Lync analytics when you walk out that door this afternoon our speakers entire career has been spent in integration of advanced analytics to solve business problems this has included his early career in quantitative analysis for a hedge fund business to using his vision of applied analytics and several startup ventures to his position is the president of Aspen analytics he co-founded Lync analytics in 2010 because he had a vision that analytics was moving towards a stronger emphasis on computer programming and mathematics as those data sets became increasingly larger building on that vision link is truly established itself as a strong player in the revolution of big data analytics and is helping businesses work through the translation of complex unstructured big data turning that into information to help improve business solutions he also has two degrees from my alma mater Georgia Tech and Georgia State University please help me in welcoming dr. will Hakes Thank You Jennifer and I want to say thanks to SAS for inviting me here today this is always an amazing production that SAS does I'm as a home theater enthusiast I'm impressed with all the technology what I wouldn't give to have some of this actually in my house and I also want to say thanks to Jennifer Jennifer is actually a friend of mine and and I wouldn't be here today without her and and I actually mean that sincerely she and I she's very humble she I went to grad school together and she's a lot smarter than I am and but I'm smart enough to know how to find really smart people part of my job but I did it back then and thanks to Jennifer for actually helping me through grad school thanks Jennifer oh by the way I wanted to also remind you about this afternoon at 1:45 the keynote address dr. goodnight and some other folks from SAS are going to be talking about high performance analytics encourage you to attend always very compelling to listen to dr. goodnight so I want to talk first a little bit about link analytics just so you know who we are lots of folks probably don't know us so we call ourselves answers from Big Data then we got a combination of services that we provide but our key which we think is a big part of where the industry is headed is in terms of building analytical technology and actually we brand ourselves in the sense that our vertical is data our vertical is communications data so the data you see in front of you is where our strength is right now actually in mobile and we provide some solutions to mobile providers today but also any type of data where things are talking to things a tremendous amount of interaction whether it's your phones that are actually sending signals right now whether you're tweeting right now whether you're talking whether you're on your iPad or your computer or whatever the case may be things talking to things but that also happens inside of cars telematics right information going back and forth from your car from some point telling someone something about you and your behavior lots of uses that type of information also in the energy world especially out on the west coast smart grid smart meter smart appliances helps companies understand behavior so they can ultimately tailor offerings for the utility customers and then ultimately we do do some work for the government well but we can't really talk about that so so this is a little bit of what what we mean what it means to us to talk about analytical technology and without going into all the details about specifically who we are and who we have on board I thought it would be helpful to classify how we see analytical technology scalability so you got good marketers and Client Services folks because no matter how good your stuff is you have to deliver it you got to convince businesses that you know what you're talking about and that your solutions fit their needs we've got good data scientists and programmers and we'll talk about what that means in just a second we've got a good applied statisticians analysts but one of our keys is down there and the on the bottom left of the screen the more hardcore mathematicians computer scientists graph theorists folks that don't really look at the world in terms of rows and columns and it's the intersection of those folks that helped us build what we think are some pretty compelling solutions so our tools well sadly that's where I fit if I must be honest with you PowerPoint are the tools of the marketers and the client services folks for the data sciences folks lots of terms that many folks in here can relate to Hadoop mateesah green plum Teradata you name it a lot of those companies are right in the middle of big data world and we've got scientists on board to deal with that our analysts and statisticians much more classically trained out of some of the applied stats programs that some of the folks are talking to today SAS based enterprise miner it's the standard it's been the standard for a long time it's going to continue to be the standard but also our more hardcore mathematicians and computer sciences folks they write their own stuff most of them were not trained in any kind of software they write C++ c-sharp Python that lets us take data in whatever form that data needs to be and find ways to provide specific answers to very very high high value questions from our clients so one of our missions inside of a small company where only a couple of years old and one of my missions as kind of a researcher in this area is to find ways to get people excited about big data and big data analytics and and that's not necessarily the folks in this room we're kind of already here we're already excited about it our mission is to try to get the people with the budgets the people with the influence the marketers the executives to get really fired up about the type of analytics that's going on in industry not just at Linc but all across this room and all across the country so we thought we put together something pretty any way we think is pretty interesting to get people fired up about big data analytics you you you you when I saw that I was pretty fired up actually then again I get I get pretty fired up about these types of things I do want to thank I don't have it anywhere actually I think in the slides but Nebo web is the agency that helped us sort of take our vision about doing this kind of thing and putting it putting it in in on screen very very powerful stuff thanks to them so we talked about this concept of data agents and in this Big Data revolution big data analytics revolution a lot of folks talk about is it height and a lot of people think that it is height I don't think that it is it can be height it can be hype if you focus on the wrong thing but if you focus on what can you do with this information with all that's out there it can change the landscape of businesses so one of the things that we did in our research it's helpful to talk to folks all the time so we don't walk around just saying what do we think about big data analytics we've conducted hundreds of what we call in-depth interviews with a research partner where we talked to business leaders all over the country for half an hour an hour and transcribe that information try to figure out what are folks saying about where we're headed from here where do they think we are and where are we headed a lot of folks in the room probably everybody knows dr. Michael rapa at NC State right the institute for advanced analytics and he had a great concept great saying everything we do generates data and so as I was thinking about that it led me to believe gosh imagine all the data that we do create in a day it's exhausting walk around all morning all afternoon all evening and even while you're asleep generating data so let me start with an example so here I am I wake up in the morning and data is already being generated why is it being generated I wake up to Pandora Pandora is on the phone so already a little bit of interaction with some software companies that have a sense for what I'm interested in now number one I can't believe that I'm actually going to show these pictures in front of an audience so I've put hands over my head but more importantly I can't believe I'm awake I can't believe I'm interacting with these companies already so I jump back in bed I hop on email I check Facebook I do all the things that we do in the morning so boom boom boom you start to think about all the companies that we're starting to interact with so there I am multitasker brushing my teeth checking sports on the iPad I'm now interacting with my my ISP interacting with websites advertisers kind of understand what my behavior is they probably know my IP address where I am the types of things I'm interested in now I'm on the scale I can't believe that I'm 15 pounds heavier than I want to be it's very frustrating to me but more specifically it's about data interaction so it doesn't always happen on all of our scales but there healthcare companies linked to smart devices inside of our scales that instantly check a heck of a lot more than your weight so now I not only know about my weight I know about a lot of other things and my healthcare providers know that information about me as well so here I am unloading the dishwasher in what is really a pair of catastrophic pajamas but that's kind of the point and this is not your ordinary dishwasher again for those of you out on the west coast this is a smart dishwasher connected to a smart meter that's connected to a smart grid and so what does that mean for us well it doesn't mean a lot yet but it will for those on the west coast I get a reduction in my bill why because you know what sometimes energy is expensive and sometimes it's cheap so if I Annie in with the utility companies and say I tell you what turn on my dishwasher between 10:00 and 4:00 a.m. I get a discount on my bill so here my kids are on the bus somewhere in the in the pile there and what I also know about that technology helps me understand that my kids are actually at school my daughter fifth grade has a smartphone and although I don't allows you to use it at school I know she's there because I've got to track me app that's sitting on her phone so I know she's at school she's safe as long as somebody hasn't actually stolen the device and my son who's three well he's I guess not valuable enough yet to be tracked but they're there near one another so here I am headed to the gym log into the gym lots of people know I'm there Jim keeps track of that I'm sure they've got a running tally of how often or actually how infrequently I'm at the gym unfortunately these days but nonetheless they know it so there I am again multitasking trying to wear my brightest workout colors interacting with technology but now I'm on a different ISP inside that gym so I got another company that knows what I'm doing working with technology made as ugly a face as I could make trying to trying to be pretty for the camera man now here I am at the coffee shop right so more isps interacting with different ISPs than my prior providers and technology now so now I've got payment technology other companies that I'm interacting with for these transactions near-field apps bump apps whatever you want to call them different technology that lets me interact with the coffee shop so I've got financial services companies that now know where I am what I'm buying and I've also got coffee shop octane in Atlanta by the way if you've never been there it's fantastic so here I am again on my iPad debating on what I'm going to do for the day sitting there trying to wonder what's what's actually happening and they're in the background if you can see that guy's computer on the right Ike hacking code to me so I think unfortunately I'm losing privacy and my my communications now I'm in my car I'm headed to a meeting and Google Maps just isn't working for me I'm trying to figure out what to do so instead I hop on my iPad which I know we're not supposed to do but we all do it so I've got multiple devices work in multiple people that I'm communicating with trying to figure out where I'm going and of course next the red light camera fortunately the red light camera doesn't know that I'm interacting with my technology but they do know that I'm getting ready to run that red light so now I got the municipality where I am in downtown Atlanta the knows where I am now I've got the camera at the office watches me walk in of course I'm smart enough to recognize that and point at them and Here I am at work finally sitting at my desk I'm a few hours into my day and I'm thinking to myself gosh think about all the companies that have already started to interact with just the companies and from a first-order standpoint I've gotten multiple communications companies some via Wi-Fi some via other connections on my iPad I got software providers app providers I got different websites I've got a municipality there's probably getting ready to write me a ticket I got different security companies I got a coffee shop I got a utility company and I probably got a hacker as well but that's beside the point second order its countless its countless how many companies I've already interacted with via ad networks via countless third-party providers that have all of these relationships that yes we've all agreed to in our user agreements already whether we like it or not and we've generated a massive amount of data how much I don't really know but it's a lot it's a lot of data thus far and I'm exhausted I can't believe how much interaction I've had and it's only 9:00 a.m. and I can't I can't do this anymore so is this data agent problem is this a big data problem I don't think so it's actually a big data analytics problem the storage exists out there and we'll talk about that in just a minute but big data analytics and all the information that we create as data agents in fact it changes our expectations it changes what we expect to come from all the companies that we interact with it's this kind of social bargain where hey if you're going to get all my information and I kind of know you're getting all my information what are you going to give back to me in return and so conversely it changes what companies are tasked with providing back to us so they're all in sort of rapid-fire stage investing in technology trying to figure out how do I use this information to serve people based on their expectations so what's Big Data is it this we've seen it all we've all seen volume velocity variety you know what I don't think so I don't like that definition I like definition from one of the folks that we talked to it Intel it's a much simpler definition but I think it's right big data is a rallying cry I don't need to put precise terms on big data big data to you you know to us we get 20 billion records a day through our data center partners with Nativa and others we have actually literally massive amounts of data coming through our data centers but I talked to some folks here in the utilities industries and they asked the question do we have big data and I say well you tell me that depends have you ever dealt with that much data before no do you know how to deal with that data no well then you've got big data so Christians another folk another guy we talked to from from Siena to network it's in the network space said you know what the storage is there the volumes cheap the next big question is what are you going to do with that data so we talked to a few folks to try to get their perspective on the differences between big data and big data analytics lots of different perspectives out there and what's interesting perhaps it was a biased sample that I talked to people weren't anywhere near as fired up about big data but they were really fired up about big data analytics so we talked to Andy from State Farm super smart guy very talented very very analytical I had no idea how analytical they were as a company so big data to him it means different things to different people but when I said hey what's big data analytics to you ah now that is something different so he makes the point and you'll see this throughout the theme you know what big data analytics to us is where our classical techniques maybe don't work so well dr. berry super smart guy at the University of Tennessee also runs the Institute for machine learning there to him Big Data never having enough RAM never having the complete data set but big data analytics is trying to find needles in a haystack on orders of magnitude and it makes the problem harder not easier and last year actual another friend of mine a guy named Jim head who's the EVP of analytics at BBDO s big agency huge analytical infrastructure to him big data it's about complexity its complexity that they're not used to seeing but in terms of big data analytics traditional inferences no longer estimated they're observed and that's a powerful statement so where are we today with big data analytics well the gap according to dr. rap at NC State the gap is largely a talent gap he thinks there's simply not enough talent out there by the way that's good for lots of folks in this room it's good if you're employed if you're a data scientist if you're working you're doing your thing if you're an employer I hope you're taking care of your employees because everybody wants to hire them McHale runs analytics at t-mobile business does are still acting too much on gut instinct mid-level management doesn't really understand this analytics as well and so big data analytics initiatives tend to fail this is not describing t-mobile specifically this is describing all the businesses that this analytics person has worked with last so3 professor at UNC they take a little bit of a different focus on analytics they're really trying to make MBAs much more technical very very valuable concept firms seem to be bogged down with investment in IT instead of spending time going through their data so where are we we're here we're in the first inning got a long way to go but that's a good thing having a long way to go is a problem but it's also an opportunity for most of us in this room so I want to talk about a few things where that describe I think where we're headed and why we're headed there so number one analytics is disruptive all models are wrong some are useful classical kind of statement from a classical statistician but if you think about where we are today all models are wrong and increasingly you can succeed without them now we can debate the merits of that last statement in fact if you look at a recent Harvard Business Review article where they talk about this type of thing for those of you that may have seen it where they debate you know we got a ton of data and if I've got a ton of data and math why do I need science I don't need science anymore I can simply observe and change and in exposing that type of concept to some of the research researchers that we talked about specifically dr. berry at university Tennessee he said you know what I think that's got it all wrong no offense don't you want to understand why the role of science in the midst of this still lets us know understand why without that role what do we have how do we improve how do we innovate if we simply rely on the data and only the math so years ago in terms of a revolution think about the internet and how it revolutionized the interconnectedness of people well today big data and big data analytics revolutionizes the connectedness of data when I think about where my career in analytics after the hedge fund were really began it was working for some of the local telecommunications company and we're building predictive logistic regression models trying to predict who's going to buy a long distance plan for example and I think about the data set that we were working with it was some third-party data that told you age and income and some of those things and you never really had exciting models they never really did anything different than the prior model but you kept rebuilding him and rebuilding them and trying to do something better but you didn't have a lot of other data to connect it to so the power it wasn't the problem with the logistic regression it wasn't the problem with the researcher it was the problem that the data didn't exist to do much more so in terms of disruption what do I mean by that big data analytics in the analytics in general is disruptive it changes the problems fundamentally it changes it give you specific examples in specific in degree industries think about market research think about market research calling folks on the phone hey what do you think about this do you like this brand did you what's your communication service provider today that type of info Meishan is being replaced by big data why ask someone their behavior when I can actually just observe it and also think about the concept of a/b testing that we mentioned earlier with respect to Google we used to think about Big Brother and big brother who's watching us all the time and all the Big Brother cameras think about George Orwell well it's actually going a bit further than that it's not just Big Brother watching us it's everybody everybody is watching what we do that changes the types of solutions that people are trying to build to serve what our needs are again in that sort of social exchange connecting data inside and outside of the firm provides huge value for those of you who were here yesterday in Tim's talk he talked about forecasting problems and they're doing some amazing stuff it was mind-boggling I approached him afterwards and thought wow that commercial for those of you saw it makes me want to leave link analytics and go work for Dow Chemical but that's an aside anyway one of the things that he talked about was the complexity in their models and part of that complexity and it's not necessarily huge data as he said but it's an increased level of complexity because they're bringing more and more data from outside sources not just macroeconomic stuff but they're bringing stuff from Nielsen and other companies in that changes the complexity of the problem number one it makes things a bit more computationally difficult number two I've now got external providers who were right in the middle of my value chain to produce really good analytics so now I got to think about that can I rely on this data source I know it's predicted this month but is it stable what if it goes away what am I going to tell management when next month I'm supposed to forecast demand I don't have a forecast because that data is no longer there connecting data is very very powerful presents its own challenges however so the rise of the chief analytics officer organization I think must come fortunately we just heard from a chief analytics officer at at teradata it's great to see that some companies are getting on board with this we've got my partner in crime my co-founder Sean is our chief analytics officer we think this is necessary to handle where analytics is headed so in general we need to create a new department a new analytics head to serve as an internal Consulting Group sit with clients understand their business needs and ensure analytical integration firms need a big data analytics Charter I think sri is absolutely dead on if you don't elevate this type of thing to the c-level and I would argue the CEO level the promise is often going to fall short so you've got you know organizational behavior may not be something that we wear analytics folks in general we didn't study OB but that kind of thing matters the structure of an organization matters in terms of whether that organization can make an impact inside the broader organization so think about the IT revolution and what IT did inside a company's not only launched a bunch of firms but it forever changed organizations we've got CIOs CTOs and massive organizational infrastructures that are they're able to handle IT and in fact when this happened think about many many many years ago this was not a core competency of most companies this was not what that company whatever that company is was in business to do but they had to do it anyway because it became critical for them doing business as usual and doing business better than their competition so they had to build the competency it wasn't a choice now what we've got inside of a lot of organization from my observation and the observation of a lot of folks that we've talked to is a lot of disagreements between different departments inside a big companies why do they disagree on analytics I'll tell you one very basic thing analytics is really valuable and so there's this jockeying for position inside of many many big companies midsize companies small companies to who's going to own analytics well I'm the data guy so I should own analytics no but I own the hardware so I'm the one that should own analytics actually I'm the chief strategy officer strategy officer so in fact the analytics ought to reside in my organization that struggle is inefficient so the chief analytics organization must rise they've got to articulate the vision both upward to the CEO and below because if folks don't get on board it simply doesn't resonate it doesn't drive change it also takes a special organization to find and foster talent I've been hiring analytics folks for a long time we as a class are very difficult people to manage no offense we're very ambitious we work really hard we expect to be rewarded we expect to move up the chain because deep down inside even though we try not to be overconfident about it each of us in the analytics world knows precisely the value of the analytics that we've got inside here we know the organizational change that it can drive that's tough to manage it's tough to manage especially if you don't have an organization equipped with the leadership skills and leadership knowledge of analytics to bring in talent and make sure that talent moves up through the food chain we've also got to embed analytics into the corporate DNA and if the chief analytics officer isn't there to do it who will and lastly we hear this quite a bit actually a very common theme in the inner interviews that we did we've got to drive partnership between business and academia a lot of businesses will tell you and I've been guilty of it myself look you keep giving me these folks out of grad school they don't have any real-world experience they're dealing with these data sets with a couple of thousand rows and maybe 50 columns that's not what my data looks like and I understand that I'm in that situation every day but what I can tell you is if the business leaders don't give them or find a way to give them the data how in the world can the universities be producing students that precisely fit what the businesses are asking for so in a chief analytics officer organization that partnership we think can flourish number three a talent shift in our universities and I suspect a lot of folks here can attest to that all right here's some bad news we academics in general and being asked to train our students and skills that often we simply don't have as dr. Priestley said my granny she's very smart don't want to detract from that but that's what they're being asked to do the smartest person in the room is the one that can communicate with the most clarity universities are not well configured to deliver the soft skills to their technical students and according to dr. Rapa he says you know what we the universities we kind of get it I mean we understand that we're we're not equipped to train people in these software skills we know we need to we just don't really think that we know how but got some good news for each undergrad give you a specific example here it's one that I actually had not thought about that's the beauty of these types of interviews dr. berry at the University of Tennessee says you know we've got a specific high demand concept here at our school it's the intersection of computer sciences and law it's big data analytics in an adversarial relationship and he's got firms coming him to sit coming to him saying give me every single one of these people I need them bring me more of them dr. rapa can attest demand for our graduates has never been higher even through the bad economy they've got absolutely no problem placing folks within a few years our MBAs quant experience tend to get exposure into the CXO office great news in fact what I might argue is that our presidential candidates probably ought to be putting money in the analytics programs both of them because we could do a lot of great things in terms of putting jobs into the economy with more analytics so in terms of talent it's a rare consensus in academia there's this shift as we've talked about a little bit to more computational skills more math more computer sciences more hardcore programming but ironically it's also a shift because businesses are demain ending give me somebody who can talk walk and chew gum at the same time give me somebody who I can put in a meeting who's not just a translation engine but give me somebody who understands technically the things that we're trying to get done as a company but also understands my business and knows how to talk to the non-technical folks and I think what you're going to see going forward you're going to see a couple of different paths you're going to see the more technical schools producing more technical people trying to impart softer skills on those folks and then you're also going to see this shift which we're starting to see in the MBA programs that tend to have more soft skills anyway those folks are getting more technical skills again we mentioned this earlier the unemployment rate with almost everybody I talked to every professor that I've talked to every person in every business that I've talked to bring me more analytics folks I will hire and I can attest to that firsthand real-world internships to hit the ground running it Kennesaw State is a great example of this we do a fair amount of work with them we've hired seven of their their graduates they and many other schools are trying to look for internships where they can place folks during school full-time internships where they're getting real work experience and sure that has certain dilemmas with respect to privacy but that's how they're getting real work experience today so when you hire somebody who's got two three semesters of experience with your company that's very very valuable and if you're me cheap new hire to bring in more specific courses that will become organized in different verticals so for example data mining at the University of Tennessee they don't use software if you're going to sign up for the data mining course the University of Tennessee you got a program the bad news is if I was there I wouldn't get in because I'm not much of a programmer I'm not really particularly skilled at C++ Python and so forth if you can't do that you can't get into their data mining program not all data mining programs are that way and I think what we're going to see over the next few years data mining programs for biostatistics we already see some of it at universities debt data mining for MBA students data mining for biostatistics data mining for business data for engineering more and more you'll see these types of courses differentiate based upon the type of talent that it can foster last number for now you're never really supposed to have four things you're really supposed to have three or five but it occurred to me that I actually only had four so here we go fourth and final thing a new ecosystem of companies emerges and I think one of the most powerful things if you think about the video that we did earlier I don't think it's an understatement to say hundreds of thousands or millions of new companies are going to jump in and take advantage of this space a whole new ecosystem emerges these are companies you know right smack dab in the middle you got sass you got a bunch of the hardware big data types of companies you got cloud era you got green plum you got new tizi you got a bunch of those companies that we're all familiar with you got a bunch of the Big Eight you know not not agency is the big thought leaders you got Accenture Deloitte IBM those folks we know all of them what about these guys most of these companies are companies you've never heard of you don't know these companies and you may never know him they're busy doing their own thing for their specific customers and as we started to put this slide together unfortunately some of the folks that that I was working with said you know we'll I think we can put together hundreds of these types of slides but we only put together one I don't want to bore you so in terms of our ecosystem why is it going to change why are millions of new company is going to emerge number one the concept of the accidental data company what do I mean by that data companies are in business they didn't intend to be data companies they didn't start as data companies they didn't want to be a data companies but they find themselves as data companies so think about Equifax Experian some of those types of companies they used to do credit scoring and some of the information they provide math seven massive amounts of data now trying to figure out how to monetize that data but more specifically think about think about the guy that wrote Angry Birds the guy that built it I can't remember his name do you think when he was building it before dreaming of making millions and selling out to Zynga do you think he thought that that was going to be an amazing way to create data no he was building something that he thought was going to be really cool and he did it and what is it now it's an absolutely massive data creation engine to understand what people are doing where they're doing it and when they're doing it think about Facebook I don't know Mark Zuckerberg personally but I suspect when he decided to build Facebook he thought about a social networking company he didn't think about a data company but what are they sitting on now they talked to one out of every seven people in the world they're a data company some of these companies are equipped to deal with data most of them are not if you're if you didn't found yourself as a data company you probably don't have the infrastructure to make sense of that data so hundreds of thousands of companies will emerge to partner with those companies to figure out how to make sense of that data just like the AI T spawn IT revolution spawned millions of companies analytics is going to do the same but there's also the production production ilysm of analytics so what do I mean by that if you think about high velocity type of data complex data whatever your environment is you can't really extract it it's tough to sample from it pull it out of here do something with it figure it out and then put it back in over here it doesn't really work that way that's not the big data analytics environment it's hard to provide classical sorts of consulting services on big data so what we're in business to do and other companies are in business to do is to find very high value questions niche data and produce answers to those specific questions in order to do that you've got to have technology if you're working with a telecom provider you're working with again ad network you're with them working with anybody that's got an online presence you have massive amounts of data on your hand in order to make sense of that you've got to build technology to do something with that so analytics as products I think has only just begun and last data itself I think becomes a vertical in when you're a small company you spend time number one trying to pay the bills but you also think about how you're going to build your brand and it occurred to me as we started to grow what our specialization was it wasn't that I knew the utility business specifically well not that I know the automotive business pretty well I might know the mobile business reasonably well but it's data its information and how to turn that information into insights it has a company we know extremely well so branding ourselves sort of communications data that type of information becomes its own vertical because that type of data if you think about knowing what somebody's doing in their car that helps high-risk financing companies decide hey do I want to give a loan to this person or insurance companies is this a good return on my investment you think about that type of data that type of underlying data vertical that serves lots of classical verticals today so I would be remiss without any recommendations just to go through a few of them for client-side types of firms you've got to advocate and brace for a move to a chief analytics officer organization yes advocate but I mean sincerely brace for it it's not easy it's pushing the rock uphill it's pushing a giant rock uphill in most organizations you got to understand that the consumption of Big Data solutions it's not the same it's harder than consuming small data solutions if you will and sure you're going to ask companies like mine and others to provide them the same way you use to get your stuff but if you're going to really make sense of it it doesn't work that that way it answers different questions and it answers them very very differently so you've got to brace yourself to consume big data solutions you got a partner with academia to drive innovation I mentioned this a little bit earlier but if you want talent to come out to drive your business forward you got to work with academia find ways it's tough I don't certainly don't have a prescription for it with respect to privacy but you've got to find ways to partner with academia if you do that you can innovate and you got to allow analytical talent to flow through it's not just bringing in some analysts and maybe moving somebody up to the director level of analytics that has to permeate throughout the organization it's got to move to middle and upper management and again arguably all the way up to the c-level and I don't care if you call the chief analytics office or not doesn't really matter to me but that talent has to be driven all the way to the top or you're not going to get truthfully the dollars in the organizational change to realize the value of big data analytics for services and solutions firms and I say this in all seriousness and I put my money where my mouth is get in the game right now this is it's the beginning we're in the first inning of what we think is a big revolution and it's not going to get easier the time to get in is right now to think about what it is you want to do and you dive in and what I can tell you is there's a lot of money chasing companies that are diving into the analytics world so if you're looking to raise capital to find people to help you do that I promise you with the right idea and the right vision it's out there the product ization of analytics is key that's how you scale and truthfully for a company like us I really don't want to compete with IBM or Accenture or Deloitte they're a lot bigger than I or Mu Sigma for that matter they've got scalability models with respect to services I can't match but if I've got a niche and I'm providing answers to specific high-value questions and I've got a technology to do it and I can give them that answer over and over and over again I've got a competitive advantage I don't have to compete with those guys and last focus on the next generation of talent that talent is going to help you innovate the talents going to help you come up with new ways to think about analytics you haven't thought of before it's also going to help you get entrenched inside your clients because delivering great solutions is wonderful but you got to have really smart people around you got to have smart folks help translate your solutions into business value so in closing simple concept perhaps you don't have to capitalize on big day analytics but it's a simple question do you think you should so thank you guys I really appreciate your your attention thanks again to SAS for having me just wanted to give you a sense for where you can find us contact information and where you can find our our video on YouTube I hope you guys check it out again thank you very much oh I think questionis we can charge some questions now if you have questions for dr. Higgs yeah right here in the front do we have a mic that can come over great thank you thank you so much it was a great presentation thank you my favorite picture was the one where you're trying to lift your bicycle was a great pet name needs help yeah we talk a lot of our data and there was a presentation yesterday from Deloitte I think it was Dave Dudley he talked about having the data is one thing but also the personal data like the sensitive data about age gender how so for example your coffee shop they might not know how old you were or you know what's your birth date and things like that and I think that kind of data can be really important in giving the final push to your analytics so how do you bridge the gap when you don't know that data and getting through some light on how do we bridge the gap and it wrestle with those issues that's that is a great question and getting into big data analytics has its own possibilities and its own problems so what you find is you start to look through something you say holy cow I know that am I allowed to know that what do I do with that but also to your point what's missing so you know I it's tough to answer that specifically I think it's it's different for different organizations I think a lot of the data that people have is simply behavior and it's hard to enrich that data to the extent you'd like it with age income and other types of information like that but what I can tell you is there's companies out there that are trying there's a lot of companies out there right now without getting into the merits of it be it right or wrong that are trying to work with some companies it's all about this integration right the intersection of data where you got some companies that know a lot of things about you in general your household you got other companies over here that understand the behavior of that household or the households around you and that connectedness of different companies is why a lot of companies are getting together and getting married so to speak it's a it's a tough it's a tough problem not sure if that answers it specifically first thank you for a wonderful presentation thank you second I want to ask if put your shoes our academic professor what you will do to develop the causes and pick then I'll prepare your student to be ready for the Big Data challenge it's a good question Jennifer know what I hear so what I would say is it does require a partnership with with industry because industry has technology that you probably don't have in university so at the University you probably don't have a big Hadoop infrastructure you probably don't have in the teaser box or a green plum blocks or whatever it is but it's helpful if you do so you partner with industry in terms of making sure you have the technology because as we talked about it I've spent a lot of time talking about big data analytics and how powerful it is well you do have to have the big data to so you got to have the infrastructure that comes from a partnership with with industries but I also think in general it's a partnership with companies to make sure you understand the pulse of the companies that are out there and I think the beauty of it is you know your ambitions may be noble you want to build a program that trains 100 PhDs or 500 PhDs to go back and be professors around the world that's really what you want to do is academic training but you got to have money to do that so if you connect yourself with the businesses and you build really good programs and businesses contribute to the school contribute to the infrastructure make commitments to hire X number of graduates and go ahead and put that commitment on the businesses challenge me to hire your grad students will rise to that challenge if you do that everybody wins you end up with better smarter graduates coming out you end up with more money inside your programs and you also get to solve all these other sorts of lofty academic goals that you might have as an institution so I would say it's it's listen to the businesses which may be controversial by the way but I think we have another question over here oh there's a gentleman on here I'm happy to look wherever okay how about here thanks for the talk very enlightening one of the questions that I have and I was going to ask you to comment on your expertise or observation is that a lot of times the big data kind of looks creepy from a consumer standpoint you gave an example today right no how much you weigh when you put your washer on things like that companies are obviously spending a lot of money to collect that information I wonder if you're advising or watching how as we increase our channels of observation do we make it two ways do we also employ some technology to communicate back to the customer because that I think that's what a Big Data makes the most impact right so it's not I'm just collecting information about your weight or when you do your own washing but I have some ways to say like hey you know what you're slipping on your diet right you want to drop down 5.0 by the way we ran it for three hours instead of four hours or they're clean or they're nothing but they're still wet or something like that because that's where a company can leverage big data to now know the insight but actually do something with it to develop a deeper connection with the customer sure I think I understand the question so there's a couple of things I think that using the specific example you mentioned and I'm on the scale and I'm a little bit too heavy somebody knows that information and in fact I've offered that information the question is what's the value exchange for me offering that information it's within that specific example it's explicit in the example that I gave of the dishwasher and the dishwasher turning on at some point between 10:00 to 4:00 a.m. that's a very explicit value exchange between myself and another company it's the bargain they are going to give me a reduced bill may be discounted premiums on my healthcare or whatever the case may be but you know you start to slip into challenges with respect to Social Policy because you know what if my weight goes from 15 pounds to 20 pounds and my weight goes from 20 pounds to 40 pounds too heavy guess what there's somebody who's got algorithm that says that's not so healthy for this person so maybe their premium should go up I don't know how we solve that one that's a that's a that specific example is again specific to the individual and the person that they've gone into that agreement with I think in general though with respect to privacy while you may not have a direct value exchange with I'll use the example of Pandora as I said the technology better be getting better the more you interact with it the more you let the communication services providers understand your information the more they better serve information to you back so that's not explicit it's kind of implicit in your relationship and if you don't like that relationship then you take your business elsewhere how do we wrestle with the privacy concerns I think that there will be there and there are often is there already is some pushback so you think about I mentioned this in the presentation all the user agreements that we sign and on some level probably from a legal standpoint you sign those user agreements they're going to use that information as they please but that's a legal standpoint that's not a business standpoint so what companies are wrestling with now is yeah technically they sign this agreement and I can use the information as I want but I guarantee you I don't know what's what's the rate at which people read these user agreements how many do you read nowadays I mean maybe yeah I don't know I you know think about cookies five seven years ago I'd go in every few days and wipe my cookies clean because I didn't want anybody to know where I was surfing now it's just not something I do so as the information that we generate gets gets stronger and stronger and more related to us I think what's happening is our tolerance for that also continues to creep up but there's probably a breaking point well I think we have a question over here on the right I really enjoyed slide when we're bulking up with your arm curls thank you but the one that's two votes for the slide Thanks the one that spoke to me the most was Ben slide with the quote from dr. box about all the models are wrong but some are useful and in the lower right corner you said it's important someone from the googles that it's enough to observe we do not need science can you expand on what it means to observe doesn't the sun's train our minds to understand what is important well III think that's a good question I guess you know how do you find that the differences between science and math I think science oftentimes is more of hypotheses trying to understand what you believe to be true and then testing whether or not those hypotheses are true versus in the case of Google they're observing something happening whether it's search or whatever the case may be and they design this specific treatment for this group and this specific treatment for that group and when this thing works better they do this thing so it's not that that's devoid of all science I mean clearly there's a scientific component to that you're trying to test something that works and if it works well then I'm going to do that more often but but I want to go back to what dr. berry at the University of C said when I talked about this concept with him and that Harvard Business Review article don't you want to understand why do you want to understand why and and what I would say is maybe I don't think you always want to understand why and far be it for me to tell Google you know I think you need to understand why these search terms are working so therefore don't optimize this in every instant as you're able to do so I think there's lots of companies that might be able to capitalize on that without the benefit of understanding why I think dr. Berry's point is you know let's not go too far there's a lot of companies who ought to be in the business of understanding why so it's it's really just a perspective I can't say that Google's perspective is wrong for Google I just think it's probably not the right perspective for everybody I think there's a question in the back back in the 2000s 2000 2002 there is a big push for privacy advocates and companies someone to work alongside the IT side or marketing side that seems to kind of evaporated and with the push now to social media and the collection of streaming data personal data personalized data do you see that coming back or just a blending of existing roles I hope so because I think what happens and we spend time talking to you know some of our clients are pretty big companies and they got big legal departments but it oftentimes seems like there's a missing person at the table when we're talking to the legal group about can we use this information or shouldn't we and what are the implications of using it there's a seat that's missing and I think the seat internally often times your point this chief privacy officer who probably has a pretty good relationship with the legal folks who understands what outsiders are trying to do for organizations but also has its pulse has his or her pulse on what's happening in the marketplace III think maybe companies have for a while seeing the value of a lot of this information and potentially you've just turned a blind eye to it but again I although I think our tolerance does continue to rise in terms of what we're willing to put up with what information we're willing to just kind of hand out to everybody all day every day it's not rising as fast as the information itself has created and so if there's a distance between those two at some point I think there's a cliff there's there's a breaking point and it's happened in other countries where you've got to sign user agreements when you go on in order to agree to be cookie so I think that will continue to push forward in terms of our where we push the privacy envelope I think there will be pauses and hopefully I think some of those pauses may be more of a return to the chief chief privacy officer type of environment you may have addressed us already but I know there's some companies out there that they're developing data markets and do you have any personal insight into that and do you see that as part of the push back to for consumers to have some means to assert their rights in the marketplace sure I understand the concept pretty well and and in concept it's an awesome idea you put together all this information you know all these things about all these people and you can do so many different things for them to customize communications or product offerings or whatever the case may be but those people don't know they're a part of that market and I see I see some pushback by the public as that type of thing becomes more mainstream I don't think a lot of people really realize that in fact I don't think a lot of people really understand how much information companies have about us today so sort of connecting the dots back there between the gentleman behind you I think that that's a great idea but I think it's one of those things where somebody said hey we can do this let's just get all this together without necessarily balancing yet or bringing the folks on board to say are we going to go to jail if we do this are we going to get in trouble are our customers going to run for the hills if they find out that we're doing this so I think there's some legal gray area there but I think the business also has to get smart to say you know what I think this can hurt our brand this is going too far let's pull back just a little bit I think we have time for one more question dr. Hicks thank you um I wanted to ask you said that there will be difficulties in trying to put a chief analytics officer in that top c-suite I wonder what you felt some of those difficulties specifically will be is it the pushback to adoption to that sort of structure what's the pushback about why is it not a layup well well I'll start with with basic concept I think big data analytics is powerful I think it's revolutionary but just because I say it doesn't make it true and I can tell that to whoever is a CEO in the audience here anybody that I'm talking to it doesn't mean that they believe me they've got to get it for themselves himself or herself the CEO if that's going to happen the CEO has to buy it you're going to change the architecture of your C suite the CEO the board and others are going to have to buy and that's not an easy thing yesterday the CEO didn't think that analytics was the key to changing his or her business and if he or she is going to tomorrow it's going to take a lot more of those crazy videos we produce to try to try to convince them so that's number one it's tough to get the CEO on board number two next to the CEO there's a CIO and a CTO and the CSO or maybe a CMO or a CDO or choose your C those folks are all jockeying for position in this analytic space it's powerful and what are we my power for well there's more of a buzz about it and guess what there's a lot of money internally you got a project you're do invest in analytics you're going to do this thing whatever this thing is and we're going to have this massive return on it I want to own that thing and if the chief analytics officer comes in then he or she may take that away from me and I don't mean to sound territorial or to negative but those are the reality of the way a lot of a lot of it times it works inside of big companies mid-sized companies as well there's the struggle for scarce resources there's a struggle to make sure think about we can measure more and more things more and more verticals are getting measured what's the return on your vertical to my business overall and if you're taking this high value analytics away from my vertical and you're now going to vertical eyes it somewhere else in the CA organization I'm not sure that's good for my organization so I think those are just a just a couple of challenges that are that are worth noting great well I think that concludes this section of the conference we really appreciate your time you did a great job thank you so much thank you for such great questions we're going to take a break now

33 Comments

  1. lenin christ said:

    nice https://bit.ly/2w8mrMC

    May 22, 2019
    Reply
  2. Urge said:

    keep it up, i dig this!

    May 22, 2019
    Reply
  3. SteveHovland said:

    What gets budget owners excited is what you can do for their top and bottom lines. Flashy video pieces mean almost nothing.

    May 22, 2019
    Reply
  4. MF Wong said:

    Very informative video. Can I embed it in my website – www.bigdataspace.net? Thanks.

    May 22, 2019
    Reply
  5. mostafa youssef said:

    I am SAS engineer (Substation automation system engineer) is there any relation between these systems and your system?

    May 22, 2019
    Reply
  6. sergiojew said:

    update my buyer big data update now
    Thanks,
    Sergio C. Adino

    May 22, 2019
    Reply
  7. Omar Sharaf said:

    this is a great video i am planning on doing a MSc in data science and i will also gain SAS certification is there anything i should learn or practice before i start the course

    May 22, 2019
    Reply
  8. Sahil Chhabra said:

    Hi All,
    I want to learn This technology from basic level.Please guide me for Analytics.

    May 22, 2019
    Reply
  9. Louis Wai said:

    Great talk

    May 22, 2019
    Reply
  10. Christian Eka Saputra said:

    that was awesome video , very insightful about big data , and useful for my project right now to make internal data analytic for my company

    May 22, 2019
    Reply
  11. Z-thru Analytics said:

    Thanks for sharing this great video seminar for the revolution of web analytics. 

    May 22, 2019
    Reply
  12. Felipe Flores said:

    A typical case of the blind leading the blind. Extremely basic talk.

    May 22, 2019
    Reply
  13. Neal Westphalen said:

    I guess the women that introduced him didn't take Bart Queen's presentation skills

    May 22, 2019
    Reply
  14. Quaalude Charlie said:

    Hadoop 🙂 QC

    May 22, 2019
    Reply
  15. Goutam Nayak said:

    AWESOME……………..m heading towards it…………………..

    May 22, 2019
    Reply
  16. Nicole Boyle said:

    Great talk.

    May 22, 2019
    Reply
  17. Riki Irfan Hidayat said:

    Great video, very informative. Thanks a lot.

    May 22, 2019
    Reply
  18. Donald White said:

    insight !

    May 22, 2019
    Reply
  19. Roy Mc Calvey said:

    Wow! wish I cud get my head round this. I have little IT capabilities. Amazing I can send an E mail.

    May 22, 2019
    Reply
  20. prasanna kumar said:

    very insightful

    May 22, 2019
    Reply
  21. Suraj V.V said:

    Awesome presentation. Very Informative. This  will definitely help me in pursuing my career in  Data Mining and Analytics 

    May 22, 2019
    Reply
  22. Sandhya C said:

    Very Informative. This makes me pursue data science as my career

    May 22, 2019
    Reply
  23. IaskedReally said:

    The woman introducing him was my statistics professor at Kennesaw State University. She's awesome. William Hakes came to KSU in 2011 to speak at SAS Day where I presented. Great Speaker!

    May 22, 2019
    Reply
  24. Shiva Subramanian said:

    Can say one thing , my eyes are open for opportunities.Thanks Will.

    May 22, 2019
    Reply
  25. Vijay Shankar said:

    Thanks Will for the great lecture. I am an engineer and am presently doing an MBA. Can you suggest more avenues to learn more about big data analytics? As an engineer, I am not bad at Math and stats and as an MBA I am learning to understand business as such. I am sure over a period of time I'll need to understand big data analytics and use it 🙂

    May 22, 2019
    Reply
  26. Gaurav Mishra said:

    Good and informative presentation. looking forward more big data presentations.
    Thanks

    May 22, 2019
    Reply
  27. ArchimedeanEye said:

    I agree, this may do more harm than good…. de-humanizing human behavior

    May 22, 2019
    Reply
  28. YourTubeVideoss said:

    This Video Is Fantastic Very Entertaining

    May 22, 2019
    Reply
  29. redarrowhead2 said:

    And it has great power to do good, just like any other piece of technology. There are benefits and harms to everything.

    May 22, 2019
    Reply
  30. rhrabar0004 said:

    This is terrifying. It is creating the framework for absolute tyranny if/when this technology eventually falls into the hands of an unaccountable state. Without privacy one cannot rebel or protest against authority. I embrace technology for the most part, but after watching this part of me thinks Orwell was being conservative in his warning. Everything we do is already logged, once the analytics are worked out, we might as well say checkmate.

    May 22, 2019
    Reply
  31. R EDWARDS said:

    Intriguing

    May 22, 2019
    Reply
  32. GANAPATHI RAO said:

    Good work..

    May 22, 2019
    Reply
  33. Trey Lyda said:

    Nice work! Very informative. I love the pics in the presentation! Great stuff! III

    May 22, 2019
    Reply

Leave a Reply

Your email address will not be published. Required fields are marked *