(upbeat music) >> Thank you and good morning or afternoon, everyone, depending on where you're coming to us from and welcome to today's breakout session, Fast Data, a retail industry business imperative. My name is Brent Biddulph, Global Managing Director of Retail and Super Bids here at Cloudera and today's hosts. Joining me today is our feature speaker Brian Kilcourse, Managing Partner from RSR. We'll be sharing insights and implications from recently completed research across retailers of all sizes in empirical segments. At the end of today's session I'll share a brief overview on what I personally learned from retailers and how Cloudera continues to support retail data analytic requirements, and specifically around streaming data ingest, analytics, automation for customers around the world. There really is the next step up in terms of what's happening with data analytics today. So let's get started. So I thought it'd be helpful to provide some background first on how Cloudera is supporting retail industry leaders specifically how they're leveraging Cloudera for leading practice data analytics use cases, primarily across four key business pillars and these will be very familiar to those in the industry. Personalize interactions of course plays heavily into e-commerce and marketing, whether that's developing customer profiles, understanding the omni-channel journey, moving into the merchandising line of business, focused on localizing sorbet, promotional planning, forecasting, demand forecast accuracy, then into supply chain where inventory visibility is becoming more and more critical today, whether it's around fulfillment or just understanding where your stuff is from a customer perspective. And obviously in and outbound route optimization, right now as retailers are taking control of actual delivery, whether it's to a physical store location or to the consumer. And then finally, which is pretty exciting to me as a former store operator, what's happening with physical brick and mortar right now, especially for traditional retailers. The whole re-imagining of stores right now is on fire in a lot of focus because frankly this is where fulfillment is happening, this is where customers steal 80% of revenue is driven through retail through physical brick and mortar. So right now store operations is getting more focused and I would say it probably is had in decades and a lot of it has to do of course with IoT data and analytics in the new technologies that really help drive benefits for retailers from a brick and mortars standpoint. And then finally, to wrap up before handing off to Brian, as you'll see, all of these lines of businesses are rogue, really experiencing the need for speed, fast data. So we're moving beyond just discovery analytics, things that happened five, six years ago with big data, et cetera and we're really moving into real time capabilities because that's really where the difference makers are, that's where the competitive differentiation is across all of these lines of business and these four key pillars within retail. The dependency on fast data is evident, it's something that we all read in terms of those that are students of the industry if you will, that we're all focused on in terms of bringing value to the individual lines of business but more importantly to the overall enterprise. So without further ado, I really want to have Brian speak here as a third party analyst. He's close in touch with what's going on retail talking to all the solution providers, all the key retailers about what's important, what's on their plate, what are they focusing on right now in terms of fast data and how that could potentially make a difference for them going forward. So Brian off to you. >> Well, thanks, Brent. I appreciate the introduction. And I was thinking as you were talking, what is fast data? Well, fast data is fast data, it's stuff that comes at you very quickly. When I think about the decision cycles in retail, they were time phased and there was a time when we could only make a decision perhaps once a month and then met once a week and then once a day, and then intraday. Fast data is data that's coming at you in something approaching real time and we'll explain why that's important in just a second. But first I want to share with you just a little bit about RSR. We've been in business now for 14 years and what we do is we studied the business use cases that drive the adoption of technology in retail. We come from the retail industry. I was a retail technologist my entire working life and so we started this company. So I have a built-in bias of course, and that is that the difference between the winners in the retail world and in fact in the entire business world and everybody else is how they value the strategic importance of information, and really that's where the battle is being fought today. We'll talk a little bit about that. So anyway, one other thing about RSR Research, our research is free to the entire world. We don't have a paywall that you have to get behind, all you have to do is sign into our website, identify yourself and all of our research, including these two reports that we're showing on the screen now are available to you and we'd love to hear your comments. So when we talk about data, there's a lot of business implications to what we're trying to do with fast data and is being driven by the real world. We saw a lot of evidence of that during the COVID pandemic in 2020, when people had to make many decisions very, very quickly, for example, a simple one, do I redirect my replenishments to store B because store A is impacted by the pandemic, those kinds of things. These two drawings are actually from a book that came out in 1997 and it was a really important book for me personally is by a guy named Steven Hegel and the name of the book was "The Adaptive Enterprise." When you think about your business model and you think about the retail business model, most of those businesses are what you see on the left. First of all, the mission of the business doesn't change much at all, it changes once in a generation or maybe once in a lifetime, but it's established quite early. And then from that point on, it's basically a wash, rinse and repeat cycle. You do the things that you do over and over and over again, year in and year out, season in and season out and the most important pieces of information that you have is the transaction data from the last cycle. So Brent knows this from his experience as a retailer, the baseline for next year's forecast is last year's performance. And this is transactional in nature, it's typically pulled from your ERP or from your best of breed solution set. On the right is where the world is really going, and before we get into the details of this, I'll just use a real example. I'm sure like me, you've watched the path of hurricanes as they go up to the Florida Coast. And one of the things you might've noticed is that there are several different possible paths. These are models and you'll hear a lot about models when you talk to people in the AI world. These are models based on lots and lots of information that they're getting from Noah and from the oceanographic people and all those kinds of folks to understand the likely path of the hurricane. Based on their analysis, the people who watch these things will choose the most likely paths and they will warn communities to lock down and do whatever they need to do. And then they see as the real hurricane progresses, they will see if it's following that path or if it's varying, it's going down a different path and based on that they will adapt to a new model. And that is what I'm talking about here. Not everything is of course is life and death as a hurricane but it's basically the same concept. What's happening is you have your internal data that you've had since this command and control model that we've mentioned on the left and you're taking an external data from the world around you and you're using that to make snap decisions or quick decisions based on what you see, what's observable on the outside. Back to my COVID example, when people were tracking the path of the pandemic through communities, they learned that customers or consumers would favor certain stores to pick up what they needed to get. So they would avoid some stores and they would favor other stores and that would cause smart retailers to redirect the replenishments on very fast cycles to those stores where the consumers are most likely to be. They also did the same thing for employees, they wanted to know where they could get their employees to service these customers, how far away were they, were they in a community that was impacted or were they relatively safe. These are the decisions that were being made in real time based on the information that they were getting from the marketplace around them. So first of all, there's a context for these decisions, there's a purpose and the bounds of the adaptive structure, and then there's a coordination of capabilities in real time and that creates an internal feedback loop, but there's also an external feedback loop. This is more of an ecosystem view and based on those two inputs what's happening internally, where your performance is internally and how your community around you is reacting to what you're providing. You make adjustments as necessary and this is the essence of the adaptive enterprise. Engineers might call this a sense and respond model, and that's where retail is going. But what's essential to that is information and information, not just about the products that you sell or the stores that you sell it in or the employees that you have on the sales floor or the number of market baskets you've completed in the day, but something much, much more. If you will, a twin, a digital twin of the physical assets of your business, all of your physical assets, the people, the products, the customers, the buildings, the rolling stock, everything, everything. And if you can create a digital equivalent of a physical thing, you can then analyze it. And if you can analyze it, you can make decisions much, much more quickly. So this is what's happening with the predict pivot based on what you see and then because it's an intrinsically more complicated model to automate decision-making where it makes sense to do so. That's pretty complicated and I talk about new data and as I said earlier, the old data is all transactional in nature, mostly about sales. Retail has been a wash in sales data for as long as I can remember, they throw most of it away but they do keep enough to create the forecast for the next business cycle. But there's all kinds of new information that they need to be thinking about and a lot of this is from the outside world and a lot of this is non-transactional in nature. So let's just take a look at some of them. Competitive information. Retailers are always interested in what the competitor is up to, what are they promoting? How well are they doing? Where are they? What kind of traffic are they generating? Sudden and significant changes in customer behaviors and sentiment, COVID is a perfect example of something that would cause this, consumers changing their behaviors very quickly. And we have the ability to observe this because in a great majority of cases nowadays, retailers have observed that customers start their shopping journey in the digital space. As a matter of fact, Google recently came out and said that 63% of all sales transactions begin in the digital domain, even if many of them end up in the store. So we have the ability to observe changes in consumer behavior, what are they looking at? When are they looking at it? How long do they spend looking at it? What else are they looking at while they're doing that? What is the outcome of them looking? Market metrics certainly, what's going on in the marketplace around you? A good example of this might be something related to a sporting event. If you've planned based on normal demand and for your store and there's a big sporting event, like a football match or a baseball game, suddenly you're going to see a spike in demand, so understanding what's going on in the market is really important. Location, demographics and psychographics. Demographics have always been important to retailers, but now we're talking about dynamic demographics. What customers or what consumers are in your market in something approaching real time. Psychographics has more to do with their attitudes, what kind of folks are in a particular marketplace, what do they think about, what do they favor, and all those kinds of interesting details. Real time environmental and social incidents, of course, I mentioned hurricanes and so that's fairly self-evident. Disruptive events, sporting events, et cetera, these are all real. And then we get the real time Internet-of-Things, these are RFID sensors, beacons, video, et cetera. There's all kinds of stuff. And this is where it really gets interesting, this is where the supply chain people will start talking about the digital twin to their physical world. If you can't say something you can't manage it and retailers want to be able to manage things in real time. So IoT along with AI analytics and the data that's generated is really, really important for them going forward. Community health, we've been talking a lot about that, the progression of the flu, et cetera, et cetera. Business schedules, commute patterns, school schedules, and weather, these are all external data that are interesting to retailers and can help them to make better operational decisions in something approaching real time. I mentioned the automation of decision-making, this is a chart from Gardner and I'd love to share with you. It's a really good one because it describes very simply what we're talking about and it also describes where the inflection of new technology happens. If you look on the left there's data, we have lots and lots of data, we're getting more data all the time. Retailers for a long time now since certainly since the seventies or eighties have been using data to describe what happened, this is the retrospective analysis that we're all very familiar with, data cubes and those kinds of things. And based on that, the human makes some decisions about what they're going to do going forward. Sometime in the not-too-distant past this data was started to be used to make diagnostic decisions, not only what happened but why did it happen? And we might think of this as, for example, if sales were depressed and for a certain product, was it because we had another product on sale that day, that's a good example of fairly straightforward diagnostics. We then move forward to what we might think of as predictive analytics and this was based on what happened in the past and why it happened in the past. This is what's likely to happen in the future. You might think of this as, for example, halo effect or the cannibalization effect of your category plans if you happen to be a grocer. And based on that, the human will make a decision as to what they need to do next. Then came along AI, and I don't want to oversell AI here. AI is a new way for us to examine lots and lots of data, particularly unstructured data. AI if I could simplify it to the next maximum extent, it essentially is a data tool that allows you to see patterns in data which might be interesting. It's very good at sifting through huge data sets of unstructured data and detecting statistically significant patterns. It gets deeper than that of course, because it uses math instead of rules. So instead of an if then or else statement that we might've used with our structured data, we use the math to detect these patterns in unstructured data and based on those we can make some models. For example, my guy in my (chuckles) just turned 70. My 70 year old man, I'm a white guy, I live in California, I have a certain income and a certain educational level. I'm likely to behave in this way based on a model, that's pretty simplistic but based on that, you can see that when another person who meets my psychographics, my demographics, my age group, my income level and all the rest, they might be expected to make a certain action. And so this is where prescriptive really comes into play. AI makes that possible. And then finally, when you start to think about moving closer to the customer or something approaching a personalized level, a one-to-one level, you suddenly find yourself in the situation of having to make not thousands of decisions but tens of millions of decisions and that's when the automation of decision-making really gets to be pretty important. So this is all interesting stuff, and I don't want to oversell it. It's exciting and it's new, it's just the latest turn of the technology screw and it allows us to use this new data to basically automate decision-making in the business in something approaching real time so that we can be much, much more responsive to real-time conditions in the marketplace. Very exciting. So I hope this is interesting. This is a piece of data from one of our recent pieces of research. This happens to be from a location analytics study we just published last week, and we asked retailers, what are the big challenges? What's been going on in the last 12 months for them, and what's likely to be happening for them in the next few years and it's just fascinating because it speaks to the need for faster decision-making. The challenges in the last 12 months are all related to COVID. First of all, fulfilling growing online demand, this is a very real time issue that we all had to deal with. But the next one was keeping forecasts in sync with changing demand and this is one of those areas where retailers are now finding themselves needing to look at that exogenous or that external data that I mentioned to you. Last year sales were not a good predictor of next year sales, they needed to look at sentiment, they needed to look at the path of the disease, they needed to look at the availability of products, alternate sourcing, global political issues, all of these things get to be pretty important and they affect the forecast. And then finally, managing the movement of the supply through the supply chain so that they could identify bottlenecks. Now, point to one of them which we can all laugh at now because it's kind of funny, it wasn't funny at the time. We ran out of toilet paper (laughs) toilet paper was a big problem. Now there is nothing quite as predictable as toilet paper, it's tied directly to the size of the population and yet we ran out. And the thing we didn't expect when the COVID pandemic hit was that people would panic and when people panic they do funny things. One of the things I do is buy up all the available toilet paper, I'm not quite sure why that happen but it did happen and it drained the supply chain. So retailers needed to be able to see that, they needed to be able to find alternative sources, they needed to be able to do those kinds of things. This gets to the issue of visibility, real-time data, fast data. Tomorrow's challenge is kind of interesting because one of the things that retailers put at the top of their list is improve inventory productivity. The reason that they are interested in this is because they will never spend as much money on anything as they will on inventory and they want the inventory to be targeted to those places where it is most likely to be consumed and not to places where it's least likely to be consumed. So this is trying to solve the issue of getting the right product at the right place at the right time to the right consumer and retailers want to improve this because the dollars are just so big. But in this complex, fast moving world that we live in today is this requires something approaching real-time visibility. They want to be able to monitor the supply chain, the DCs and the warehouses and their picking capacity. We're talking about Echo's, we're talking about Echo's level of decision-making about what's flowing through the supply chain all the way from the manufacturing door to the manufacturer through to consumption. There's two sides of the supply chain and retailers want to look at it. You'll hear retailers and people like me talk about the digital twin, this is where this really becomes important. And again, the digital twin is enabled by IoT and AI analytics. And finally, they need to increase their profitability for online fulfillment. This is a huge issue, for some grocers the volume of online orders went from less than 10% to somewhere north of 40%. And retailers did in 2020 what they needed to do to fulfill those customer orders in the year of the pandemic, that now the expectation that consumers have have been raised significantly. They now expect those features to be available to them all the time and many people really like them. Now retailers need to find out how to do it profitably and one of the first things they need to do is they need to be able to observe the process so that they can find places to optimize. This is out of our recent research and I encourage you to read it. Now when we think about the hard one wisdom that retailers have come up with we think about these things, better visibility has led to better understanding which increases their reaction time which increases their profitability. So what are the opportunities? This is the first place that you'll see something that's very common and in our research, we separate over-performers, who we call retail winners from everybody else, average and under-performers. And we've noticed throughout the life of our company that retail winners don't just do all the same things that others do, they tend to do other things and this shows up in this particular graph. This again is from the same study. So what are the opportunities to address these challenges I mentioned to you in the last slide? First of all, strategic placement of inventory throughout the supply chain to better fulfill customer needs. This is all about being able to observe the supply chain, get the inventory into a position where it can be moved quickly to fast changing demand on the consumer side. A better understanding and reacting to unplanned events that can drive a dramatic change in customer behavior. Again, this is about studying the data, analyzing the data and reacting to the data that comes before the sales transaction. So this is observing the path to purchase, observing things that are happening in the marketplace around the retailer so that they can respond very quickly, a better understanding of the dramatic changes in customer preference and path to purchase as they engage with us. One of the things we all know about consumers now is that they are in control and literally the entire planet is the assortment that's available to them. If they don't like the way they're interacting with you, they will drop you like a hot potato and go to somebody else. And what retailers fear justifiably is the default response to that is to just see if they can find it on Amazon. You don't want this to happen if you're a retailer. So we want to observe how we are interacting with consumers and how well we are meeting their needs. Optimizing omni-channel order fulfillment to improve profitability. We've already mentioned this, retailers did what they needed to do to offer new fulfillment options to consumers. Things like buy online pickup curbside, buy online pickup in-store, buy online pick up at a locker, a direct to consumer, all of those things. Retailers offer those in 2020 because the consumers demand it and needed it. So when retailers are trying to do now is to understand how to do that profitably. And finally, this is important and never goes away is the reduction of waste, shrink within the supply chain. I'm embarrassed to say that when I was a retail executive in the nineties, we were no more certain of consumer demand than anybody else was but we wanted to commit to very high service levels for some of our key categories somewhere approaching 95% and we found the best way to do that was to flood the supply chain with inventory. It sounds irresponsible now, but in those days that was a sure-fire way to make sure that the customer had what she was looking for when she looked for it. You can't do that in today's world, money is too tight and we can't have that inventory sitting around and move to the right places once we discover what the right places. We have to be able to predict, observe, and respond in something much closer to real time. Onto the next slide, the simple message here, again a difference between winners and everybody else. The messages, if you can't see it you can't manage it. And so we asked retailers to identify to what extent an AI enabled supply chain can help their company address some issues. Look at the differences here, they're shocking. Identifying network bottlenecks, this is the toilet paper story I told you about. Over half of retail winners feel that that's very important, only 19% of average and under-performers, no surprise that they're average and under-performers. Visibility into available to sell inventory anywhere within the enterprise, 58% of winners and only 32% of everybody else. And you can go on down the list but you get the just, retail winners understand that they need to be able to see their assets and something approaching real time so that they can make the best decisions possible going forward in something approaching real time. This is the world that we live in today and in order to do that you need to be able to number one, see it and number two, you need to be able to analyze it, and number three, you have to be able to make decisions based on what you saw. Just some closing observations and I hope this was interesting for you. I love talking about this stuff, you can probably tell I'm very passionate about it. But the rapid pace of change in the world today is really underscoring the importance, for example, of location intelligence as a key component of helping businesses to achieve sustainable growth, greater operational effectiveness and resilience, and ultimately your success. So this is really, really critical for retailers to understand and successfully evolving businesses need to accommodate these new consumer shopping behaviors and changes and how products are brought to the market. And in order to do that they need to be able to see people, they need to be able to see their assets, and they need to be able to see their processes in something approaching real time, and then they need to analyze it and based on what they've uncovered, they need to be able to make strategic and operational decision making very quickly. This is the new world we live in, it's a real-time world, it's a sense and respond world and it's the way forward. So Brent, I hope that was interesting for you. I really enjoyed talking about this as I said, we'd love to hear a little bit more. >> Hey, Brian, that was excellent. I always love hearing from RSR because you're so close to what retailers are talking about and the research that your company pulls together. One of the higher level research articles around fast data frankly, is the whole notion of IoT, right? Now many does a lot of work in this space. What I find fascinating based off the recent research is believe it or not, there's $1.2 trillion at stake in retail per year between now and 2025. Now, how's that possible? Well, part of it is because of the Kinsey captures not only traditional retail but also QSRs and entertainment venues, et cetera, that's considered all of retail. But it's a staggering number and it really plays to the effect that real time can have on individual enterprises, in this case we're talking of course about retail. So a staggering number and if you think about it, from streaming video to sensors, to beacons, RFID, robotics, autonomous vehicles retailers are asking today, even pizza delivery and autonomous vehicles. If you think about it, it shouldn't be that shocking, but when they were looking at 12 different industries, retail became like the number three out of 12 and there's a lot of other big industries that will be leveraging IoT in the next four years. So retailers in the past have been traditionally a little stodgy about their spend in data and analytics. I think retailers in general have got the religion that this is what it's going to take to compete in today's world, especially in a global economy and IoT really is the next frontier, which is kind of the definition of fast data. So I just wanted to share just a few examples or exemplars of retailers that are leveraging the Cloudera technology today. So now they pay for advertisement at the end of this, right? So what is Cloudera bringing to market here? So across all retail verticals, if we look at, for example, a well-known global mass virtual retailer, they're leveraging Cloudera data flow which is our solution to move data from point to point in wicked fast space. So it's open source technology that was originally developed by the NSA. So it is best to class movement of data from an ingest standpoint, but we're also able to help the round trip. So we'll pull up sensor data off all the refrigeration units for this particular retailer, they'll hit it up against the product lifecycle table, they'll understand temperature fluctuations of 10, 20 degrees based on fresh food products that are in the store, what adjustments might need to be made because frankly store operators, they'll never know refrigeration, they'll know if a cooler goes down and they'll have to react quickly, but they won't know that 10, 20 degree temperature changes have happened overnight. So this particular customer leverages further data flow to understand temperature fluctuations, the impact on the product life cycle and the roundtrip communication back to the individual department manager, let's say a produce department manager, deli manager, meat manager. Hey, you had a 20 degree drop in temperature, we suggest you lower the price on these products that we know are in that cooler for the next couple of days by 20%. So you don't have to worry about freshness issues and or potential shrink. The grocery with fresh product, if you don't sell it, you smell it, you throw it away, it's cost to the bottom line. So critically important and tremendous ROI opportunity that we're helping to enable there. From a leading global drugstore retailer, so this is more about data processing and we're excited of the recent partnership with the Nvidia. So fast data isn't always at the edge with IoT, it's also about workloads. And in retail, if you are processing your customer profiles or segmentation like intra day, you will never achieve personalization, you will never achieve one-on-one communications with retailers or with customers, and why is that? Because customers in many cases are touching your brand several times a week. So if taking you a week or longer to process your segmentation schemes, you've already lost and you'll never achieve personalization, in fact, you may offend customers by offers you might push out based on what they just bought yesterday you had no idea of it. So that's what we're really excited about, again with the computation speed that Nvidia brings to Cloudera. We're already doing this today, we've already been providing levels of exponential speed and processing data, but when Nvidia brings to the party is course GPUs right, which is another exponential improvement to processing workloads like demand forecast, customer profiles. These things need to happen behind the scenes in the back office much faster than retailers have been doing in the past. That's just the world we all live in today. And then finally, from a proximity marketing standpoint or just from an in-store operations standpoint, retailers are leveraging Cloudera today, not only data flow but also of course our compute and storage platform and ML, et cetera, to understand what's happening in store. It's almost like the metrics that we used to look at in the past in terms of conversion and traffic, all those metrics are now moving into the physical world. If you can leverage computer vision in streaming video, to understand how customers are traversing your store, how much time they're standing in front of the display, how much time they're standing in checkout line, you can now start to understand how to better merchandise the store, where the hotspots are, how to in real time improve your customer service. And from a proximity marketing standpoint, understand how to engage with the customer for right at the moment of truth, right, when they're right there in front of the particular department or category, upward leveraging mobile device. So that's the world of fast data in retail and just kind of a summary in just a few examples of how folks are leveraging Cloudera today. From an overall platform standpoint of course, Cloudera is an enterprise data platform, right? So we're helping to enable the entire data life cycle, so we're not a data warehouse, we're much more than that. So we have solutions to ingest data from the Edge, from IoT, leading practice solutions to bring it in. We also have experiences to help leverage the analytic capabilities of data engineering, data science, analytics and reporting. We're not encroaching upon the legacy solutions that many retailers have today, we're providing a platform that's open source that helps weave all this mess together that existed retail today from legacy systems because no retailer frankly is going to rip and replace a lot of stuff that they have today. Right. And the other thing the Cloudera brings to market is this whole notion of on-prem hybrid cloud and multicloud, right. So our whole culture has been built around open source technology as the company that provides most of the source code to the Apache network around all these open source technologies. We're kind of religious about open source and lack of vendor lock-in, maybe to our fault, but as a company we pull that together from a data platform standpoint so it's not a rip or replace situation. It's like helping to connect legacy systems, data and analytics, weaving that whole story together to be able to solve this whole data life cycle from beginning to end. And then finally, I want to thank everyone for joining today's session, I hope you found it informative. I can't thank Brian Kilcourse enough, like he's my trusted friend in terms of what's going on in the industry. He has much broader reach of course in talking to a lot of our partners in other technology companies out there as well. But I really appreciate everyone joining the session, and Brian, I'm going to kind of leave it open to you to any closing comments that you might have based on what we're talking about today in terms of fast data and retail. >> First of all, thank you, Brent. And this is an exciting time to be in this industry. And I'll just leave it with this. The reason that we are talking about these things is because we can, the technology has advanced remarkably in the last five years. Some of this data has been out there for a lot longer than that and it frankly wasn't even usable. But what we're really talking about is increasing the cycle time for decisions, making them go faster and faster so that we can respond to consumer expectations and delight them in ways that make us a trusted provider of their lifestyle needs. So this is really a good time to be a retailer, a real great time to be servicing the retail technology community and I'm glad to be a part of it and I'm glad to be working with you. So thank you, Brent. >> Yeah, of course, Brian. And one of the exciting things for me too, I've being in the industry as long as I have and being a former retailer is it's really exciting for me to see retailers actually spending money on data and IT for a change, right? (Brian laughs) They've all kind of come to this final pinnacle of this is what it's going to take to compete. You and I talked to a lot of colleagues, even salespeople within Cloudera, like, oh, retail, very stodgy, slow to move. That's not the case anymore. >> No. >> Everyone gets the religion of data and analytics and the value of that. And what's exciting for me to see as all this infusion of immense talent within the industry that we couldn't see years ago, Brian. I mean, retailers are like pulling people from some of the greatest tech companies out there, right? From a data science, data engineering standpoint, application developers. Retail is really getting its legs right now in terms of go to market and the leverage of data and analytics, which to me is very exciting. >> Well, you're right. I mean, I became a CIO around the time that point of sale and data warehouses were starting to happen, data cubes and all those kinds of things. And I never thought I would see a change that dramatic as the industry experience back in those days, 1989, 1990, this changed doors that, but the good news is again, as the technology is capable, we're talking about making technology and information available to retail decision-makers that consumers carry around in their purses and pockets as they're right now today. So the question is, are you going to utilize it to win or are you going to get beaten? That's really what it boils down to. >> Yeah, for sure. Hey, thanks everyone. We'll wrap up, I know we ran a little bit long, but appreciate everyone hanging in here with us. We hope you enjoyed the session. Our contact information is right there on the screen, feel free to reach out to either Brian and I. You can go to cloudera.com, we even have joint sponsored papers with RSR, you can download there as well as other eBooks, other assets that are available if you're interested. So thanks again, everyone for joining and really appreciate you taking the time today.