Speaker 1: All right. Good evening, everybody. Good evening, everybody. Okay. All right. Just checking. Just checking. Well, first of all, Mitch and Bob, thanks for the invite. Looking forward to sharing a little bit about leveraging big data analytics for small to medium-sized businesses. So what I'd like to do is give you a little bit of a baseline on what big data analytics is, talk a little bit about the hype associated with it. And there's a lot of hype around big data, but it's probably been over the last 10 years. It kind of has trended down a bit. But for small businesses, it's on the rise. Also, I want to have a little short video for you as well, just a little background on big data analytics. I like to spend most of the time on what should small businesses consider in this space. So I think when you think about big data analytics, I mean, when you look at the history, I mean, big data has been around a long time. When you think about just the government, the U.S. government, just think about all the data that they've had to manage. When you talk about just digitizing tax returns, Social Security, Medicare, things of that nature, they've been dealing with big data for a while. Now, obviously, I mean, it's not big data in the same context, right? But in the 60s, they had the plan for building the first data center in the world. So I mean, they were the first ones to kind of just think of the concept of truly big data. I wouldn't say that that's a great milestone to mention, but I mean, they were one of the early folks to think about big data. But when you fast forward a couple of decades into the 90s, I mean, you're looking at the web, right? Looking at the Internet, some of its predecessors. And, you know, I think the use case there was obviously search and obviously searching and getting results very quickly and obviously very reliably. So that was really maybe the next big wave of big data, so to speak, in the search category. I mean, you had to deal with not just the search technology, but also the infrastructure associated with it. And as you move forward, maybe the next big wave with regards to big data is probably more the mobile space, the rise of social, also just sensor data from IoT and things of that nature. So now companies have all these data sources, right? And it would be nice to kind of pull that together and try to get some, you know, get some feedback, get some things that they could use to run their business a little bit better. So that's really the concept of big data. Now, I do like this quote from the CEO of IBM. Basically, she says that big data will spell the death of the customer of customer segmentation. And people are familiar with the concept of customer segmentation, meaning that, you know, you treat groups of customers the same. She's saying that that's really going to be big data is going to be the death of that and really force companies or the marketer to actually understand each customer as an individual. And if you don't do it quickly, you're really going to be left in the dust, you know. And I think all of us being consumers, we kind of understand that to a certain degree. I mean, if you have a Netflix account, so to speak, or Amazon, I mean, you can kind of see it's very personal, right? So the concept of customer segmentation is not going to go away very quickly. But the concept of personalization is definitely on the rise. Okay. So what is big data analytics? Now, I think at the at the highest level, really, what we're talking about is really taking large, very complex data sets, often very disparate, very, maybe some structured, you know, maybe unstructured data, bringing that together, right into a single data set. You obviously want to run a series of advanced analytics with that and hopefully uncover, you know, some some findings, some some things that are maybe unknown, to a great degree, that hopefully gets you to a point where you could actually run your business better, but actually, even more importantly, be able to be more predictive or prescriptive about something that might happen in the future that you can prevent. You know, I mean, that would be the goal, or at least a high level picture of what big data is about. I think also, I think the the big part about big data is really the people side. It's very, it's unique in the sense of when you look at the skills that are required to really leverage big data. You know, you're talking about not just IT experts, but I mean, it's really specific skills. So I mean, you really need to have some pretty solid programming skills. And I mean, really coding skills, you know, in Python, or in some of the contemporary programs. Also, you need to have some knowledge regarding the architecture and the systems that are involved, for sure, in terms of you need to move large amounts of data across systems, you also need to be able to visualize that data and actually make sense of it for your clients. And then you also need to be pretty deep. As a quantitative thinker, I mean, you really need to have some math, mathematics depth, you need to have some statistics knowledge as well. So very specific knowledge there to be able to make sense of what you're what you're analyzing. And then on top of that, you really need to be have some business savvy, you know, I mean, you get these results, but does it really make sense? Or, you know, how can the business use the results coming out of that? So it's, it's a very unique skill set when it comes to some of the people who are involved in big data analytics. And I think the big guys, I know, you mentioned that you were at Accenture, but you know, companies like Accenture, that have 1000s of people in this space, they they have it figured out. I mean, they have groups of people with those different skills collated together, co located together to actually, you know, solve client problems. The small to mid size business, that's a challenge, right? And we'll talk about maybe what they can do to kind of get those skills. Now, the technology. I think when you look at the technology, there's some basics that probably most small to mid sized business have, they might have to get boned up a little bit on some of the programming skills, but the basics, they probably do have. The next step is really the platforms and the platforms that are out there. I mean, I have some of the big guys on the slide here. But there's definitely if you take a tier two or tier three level provider, you can get some pretty solid advanced analytic skills for a small to mid size company at a very, you know, affordable price. And I'd say also, over the last maybe five to 10 years, I mean, when you look at some of the technological advances that have really helped with lowering some of the prices, particularly in the cloud space, you know, whether it's compute or data storage, that's really has made it really affordable for small to mid sized businesses to take advantage of this technology. So that's kind of a little bit of the push, you know, for small to mid sized companies to get into the space. So the hype. So I think when you think about hype, I mean, it comes down to the dollars first off, so when you look at the market for big data, it's definitely growing. I mean, I think last year it was in the 14-15% kind of growth rates. I mean, when you look at the forecast going forward, they're double digit kind of growth rates for the marketplace. So I think that's understood. More importantly is the results. And now the results are based on the fact that you need to have successful implementations, right? So, I mean, not obviously 100% of them have been successful, but of the ones that have been successful, clearly you're seeing processing costs and expense costs going down significantly. And I think in the industry that I came out of, basic materials, chemicals industry, they've been leveraging big data technologies for maybe a decade or more, depending on the vendor. And you can think of it from the standpoint, if you're making a delivery of a product to a customer, you're scheduling a product to a customer and needs to be there at a certain time, very cost-effectively, and you want to be able to improve that every year, right? So that industry or industries have been using that technology for a while, continuing to enhance it, continuing to enhance it with machine learning types of capabilities. So that industry is probably one of the early adopters. I think also when you look at I think as consumers, we kind of all understand new avenues for innovation and being disruptors. I think everybody in this room probably has an Amazon account, probably has probably a Netflix account, I would probably guess in this room. And when you think about those kinds of companies, I mean, just that algorithm that they use to recommend, I mean, that's really a big data exercise. That's really what it is. And obviously, that's proprietary to them. But clearly, that's a common use case right now for big data. And I think the last two, speed, customer experience, mitigating risk and fraud, I think those, if you look at probably the financial services industry, banking, as an example, they probably have leveraged that space, maybe not the most, but quite a bit. I mean, I kind of think of an example, I think about five, six years ago or so, I was on vacation, my family were on a vacation. I think it was in the Smoky Mountains, but don't quote me, I think it was in the Southeast. And we were out, we were at a very remote location, I tried to pay for something at a souvenir shop and my card got declined. And it never gets declined. And I called them up, they said, well, you know, Mr. Anderson, you haven't bought anything from that location or in that general region ever. And we have a little bit of a fraud alert in this particular part of that particular state. So that's why we, we canceled your, we declined your transaction. But since you called, you know, we'll remove the flag. And so, you know, I would say that's probably one of the early use cases at that particular credit card company in terms of just leveraging real-time big data analytics. And now I think all of us are very familiar with that. I think almost most of the major banks are way beyond that at this point in time. So, I mean, utilization, I mean, clearly most of the Fortune 500 companies are all over big data analytics. They would say it's a critical enabler for them to be competitive. I think if you also, I think this quote is actually from Accenture, this last one, I mean, about most executives feel that if they don't do big data analytics, they kind of feel that they're going to lose their competitive position or really be, you know, kind of like left in the dust. You know, whether that is 100% true or not, I guess we'll have to see. But that's how serious a lot of the executives at the Fortune 500 companies think of big data analytics. So how can small businesses leverage big data? And in a way, it's kind of like not rocket science. You know, I mean, you kind of have to start with strategy, right? And, you know, I think probably most of you in this room, and I think I heard in Matias' presentation this previously, when you look at failures in digital transformation or strategic transformation, the number one reason is clearly lack of clarity around their strategy and what they want to accomplish. But assuming that in this case, they're successful, I mean, what you need to do, first of all, is to really define your strategy focused on really what are the problems, what are the inefficiencies, what are the pain points, but link them to business outcomes, what you want to actually see in your P&L, in your balance sheet. And some people call that a digital strategy, digital transformation, but you really need to link that to your business strategy. So that's the very first thing. So once you have the strategy defined, the next thing is really around the right process and also the right tools to select. Now, I had a discussion actually earlier today with a CO, with a manufacturing company, and it was in the space of advanced analytics. And we were talking about, you know, what are some options, what are some of the challenges he's facing? Now, there was a particular process in the operation space. However, they were not gathering much data at all. He knew it was a problem area. So that's probably not the first place you want to actually try to do advanced analytics. You're going to want to try to have a place where you have some data or be able to gather some data and establish a baseline before you do any kind of advanced analytics. So the very first thing is really around data availability. So assuming you have that, then the next step is really around, I would say, is choosing one of the tools. And there's a variety of tools. I have one on there right now, but I've used a few that have definitely some free kind of capability for a period of time and some price points that are pretty low that I think a small, mid-sized business can cover for sure, depending on the amount of data that you're talking about, the types of analytics you'd like to do, and some of the visualizations you want to see. So there's definitely some options. You don't have to use Watson Analytics. I mean, I have it up there. It's probably a little bit of the benchmark, so to speak. But there's a lot of even local companies in the metropolitan Philadelphia area that have some capabilities in this space. Then the very next, I mean, then we're talking about execution. So we're talking about starting small, thinking big. So you're talking about actually trying to start with a proof of concept, build some momentum through quick wins, and then look at scaling from there once you have your success. But you can't really do any of the sustainability and change management stuff. You don't really wait till you actually get through the proof of concept. You really have to do some of that work at the strategy phase. I mean, you can't wait till the last minute. Otherwise, you kind of have some of the examples that Matthias mentioned in his previous presentation. So you go through the proof of concept. Then you get to the point of scaling. And then I think when you think about scaling, I think most of the people in this room are very familiar about scaling from an IT perspective, things that you need to take into account. Same for big data analytics. It's going to be some of the same challenges, same issues you're going to have to think about. Change management, I think, is also, I think it's critical in any technology transformation. I think in this space, it may be a little bit more. I mean, maybe almost as tough as RPA, artificial intelligence. But I think in this space, the people who are doing the work today, I mean, your reporting analysts, your business analysts, whether they have the skills to kind of do advanced analytics, it's going to be a challenge for them. So I mean, you're going to want to engage them early in the process. Whether they're able to move with the technology or not will be something that you'll have to find out and see. So yeah, I mean, that's pretty much the straight, that's pretty much the process. Like again, it's not rocket science per se. But it's really just the first step though, because there's a lot in the big data analytics space. You know, I mean, when you look at some of the options beyond machine learning into the cognitive space, there's many options that you can choose from an advanced analytics perspective, and this would really just be your first step. So on that note, any questions? I think baseball is going to follow. I think they're definitely going to go more into big data. I mean, they have, right? I mean, when you look back at the Oakland A's and Billy Beanball, I mean, I think they were kind of the forerunner, right? They talked about the automation of the strike zone, you know, hauling balls and strikes. I think that's going to happen in the next three years, right? And so, yeah, I think with regards to analytics, I mean, I think that's a big part of almost every single professional sport now. So I would think that that's going to continue. I mean, it's amazing. You know, I'm on the advisory board with my alma mater, and, you know, you kind of think of the staff, you know, at a college football program, you know? I mean, you have analytics interns, you know, in addition to analytics people as part of the coaching staff, right? And that's in college. So I think in the professionals, it's going to be even more. I like the quote of the IBM. I think that's definitely more B2C at the moment. I don't know if there's a lot of B2B types of use cases yet that kind of take advantage of that. But I think it's definitely B2C is all over that space. But no, I don't have like a real live use case that I've seen in either chemicals or in the B2B space. Maybe there, I'm sure there are, but I'm just not aware of any. Well, you know, I think there's two things. I think, yes, there is a threshold, but I think big data is kind of relative, you know? I think from the standpoint of, for example, I've worked with clients where you have big data that where you have maybe, you know, two to three million lines of rows of data, maybe hundreds of columns of data. I just finished an engagement that was really more like 100 to 200,000 rows of data and maybe a couple hundred columns. So, I mean, I say it's relative. I'm not trying to be facetious, but big data for a small and medium-sized customer or client is very different from a Fortune 500 kind of company. However, I will say that there are some thresholds, you know? I mean, the analytics will tell you when you have not enough data to be able to extract an answer for you, right? Yeah, correct. And, you know, and if the data is kind of all looks the same and there's not a lot of variation, I mean, the model will tell you that too. There's not a lot of, enough variation for us to be able to come up with an answer. So, I mean, there are some constraints on the model with regards to either the size of the data set, the variability of the data set, et cetera. But yeah, there are some constraints around the machine learning algorithm. You know, I think clearly the strategy of the, you don't want to blame it on one particular industry, but the strategy that was employed around what mortgages actually got approved might have been more of a contributing factor. But I think big data analytics also had a contributing part to that as well. I'm not sure about privacy. Security, for sure. You know, I think every single client has a challenge with, let's just take example of a particular use case, you know, around predictive maintenance, right? So, it's becoming common now for someone to put a device on a compressor, right? To measure vibration and other things and feed that data into the cloud. And the customer feels that that should be our data. The vendor feels like, well, that should be our data. And every single situation is always a negotiation who owns the data and the security of that data. But I haven't, I've seen that happen almost every single situation. But I haven't heard the comment regarding privacy, which I think is a good point. It's a good point. I think it's in the fine print, but I think they're more concerned about the security of the data versus the privacy, at least at this point.