 here, this is theCUBE. We're live in Boston, Massachusetts for HP Vertica's user conference. Hashtag is HP Big Data 2013. This is theCUBE, our flagship program. We go out to the events, extract the signal from the noise, and share the best use cases with you here in the big data space. Obviously HP's doing very, very well. A lot of great practitioners here. Cutting-edge customers really doing some great work in the area of big data, business intelligence, data warehousing, all the transformative value-creating propositions are happening right here with Vertica. I'm John Furrier, the founder of SiliconANGLE. I'm joined by my co-host. Hi everybody, I'm Dave Vellante of Wikibon.org. Bruce Yen is here, he's the director of business intelligence at Guess. Bruce, welcome to theCUBE. It's a pleasure to be here. So, here in an industry, fashion business, retail business, it's elegant, it's clean, it's a tough business, but fashion is simple and alluring and compelling. Can you make data sort of match that same metaphor? Wearable data, right. I think it is challenging, you know, because we have users that expect our data to be easy to use and that's actually led us to a lot of different solutions and we've done a lot on mobile BI because of that, so that's a good point. So, talk a little bit about your business, what some of the drivers are and talk a little bit about your role. So, our business, I mean, we're a global retailer, right, and we have about 1,600, over 1,600 stores in 87 countries and we are growing, going to different regions and different countries. In fact, we're opening Japan later this year and so it's really exciting. I've seen the company grow from a wholesaler, primarily, into a retailer, into an international retailer. So, my role is really to provide the business with information, with data, to make it accessible and make it easy to use and really also look at some of the emerging trends with data and figure out how we can best leverage that. Now, just I inferred from reading about some of your background that you really were handed this problem, kind of a blank sheet of paper, spreadsheets and snapshots of photos. Maybe you could take us back to the drivers of, you know, when somebody said, hey Bruce, you tapped his shoulder and said we need you to help us solve this problem. What was it like, what were the drivers and what was the situation like back then and let's take it through sort of where it is today. Right, so the history of our data warehouse initiatives, we've always had data and we've created these data warehouses originally on Oracle, now we've moved over to Vertica. And, you know, our users, they're using, you know, spreadsheets, they're running reports and at the end of the day, you always have some users that just aren't that comfortable with that and we saw it as an opportunity to combine the rich media from our best sellers and basically use the links that we have online from our e-com store to really populate all the images so that they can easily just see what the best sellers are and to us it was kind of groundbreaking because initially we all had reports with style numbers and those are really hard to memorize. There are a few that do memorize numbers which is pretty tough but really to see it visually, you see so many things that there's a host of visual analytics such as just looking at the colors and the styles and the fabrics that are actually taking shape, putting them side by side, that's huge. You know, no report with tons of written data is actually going to get that for you. Was there some friction at first from some of those guys that would like to memorize some of that stuff? No, they loved it. They felt that we were basically kind of freeing up, you know, part of their time to like really just be able to analyze the information instead of compiling it and having people put the images on spreadsheets. So what we did was then, our first project was really to create these dashboards and to us it was huge because our buyers, our merchants, you know, usually, you know, they're pressed for time and they're able now to really leverage this information and in fact, it took off. People would ask us, can I get access to it? And we're like, no, you already have access to it but they asked us in a way that we knew that they really wanted to use it and that was a very exciting course. So you had a big demand for this. So expectations were probably pretty high as well. So how did you meet or exceed those expectations? Well, I think we looked at it from our company's cultural standpoint. We're a fashion industry. We're in the fashion industry and people like pretty things, you know, and we're very close to Hollywood, you know, we're based out of LA and we wanted to make it fun and just exciting to use. So we had a graphic designer actually go in there and re-skin these dashboards. Plus we looked at the workflow to make it really easy to use. So I think that really helped the adoption and we've done that ever since. With anything that's visual, we hire graphic designers and really we make sure that they're easy to use and the usability is actually there. Bruce, what are some of the things that you're seeing? I mean, honestly, you guys have using some big data stuff for the current business and transforming that and obviously there's benefits around seeing patterns and user behavior around your customers and analytics and all kinds of stuff you can now use on the BI side. But just generally beyond that, I'll see the trends of Google Glass, Fitbit, wearable computing, devices, the instrument. You could potentially have some sensors on jeans, right? Or I'm making that up. But it's possible down the road, obviously. But this brings up the user, the consumer experience, right? Where the fashion and the elegance of clothing, obviously you're seeing that tied to phones and smart phones. So is that something that you see happening in the fashion business and what do you guys look at there? What should we take on that? I think our take is really, how do we use technology to enhance the user experience? Enhance the customer experience. When they walk into our stores, when they're online, how do we better market to them? So I think part of it is really going back to, I guess, basic retailing is, do you really know your customers? How well do you know them? And with the advent of all these amazing technologies, such as Vertica, we're able to leverage all this data. And I like to say, we have a little big data problem or big little data problem because I think for a lot of retailers, we have a lot of structured data and we deal in a lot of structured data and just trying to make sense of it. Create more relevant recommendations when you go online, figure out how to pipe those different types of analytics to all parts of the business. That's key for us. So that's interesting. So I'm thinking about when I walk into a retail store and I'm looking for something. You know, sometimes you're put off when somebody's in your face, but other times when somebody's really helpful, oh, that looks really good on you or they make good suggestions. So you're trying to essentially copy that experience but online, but it's going to be a very difficult thing to do. So how do you get that right? I mean, people always talk about A-B testing. John said it's A-B-C through Z testing these days, but maybe you could talk about how you've iterated and perfected that model. Well, I think it does go back to A-B testing and it does go back to a lot of testing because we create these models with some assumptions and just making sure that just to see if it increases lift or not, if customers are actually clicking on these recommendations. So to us, I think part of it is having business expertise in our field and for a particular brand of retail and being able to really leverage that and create these models and test them. So there is a lot of trial and error. So I don't know if you can share any specific metrics, but even just in general terms, maybe in the baseball family, baseball terms, walk, single, home run, what was the business impact of some of these innovations that you've brought to guests? Well, I think we're actually, we're still in the process of measuring how impactful the recommendations are that we actually leverage a lot of the data from in-store and actually online and incorporate that together for our analytics for recommendations. So we're still actually in the process of finding out how successful we are and actually still fine tuning it. So it's actually, it's very promising only because I see how from a company standpoint, we're beginning to look at analytics a lot more as something that's a lot more valuable. And I think for a lot of retailers, we're a 30 year old plus company, but a lot of times we still run the company like a mom and pop. We still have that DNA within our company to be innovative, to really be entrepreneur. So I think it's, in a way, that's great to have that DNA, but sometimes to be an analytics driven company, it's a little bit more difficult. So I don't know if you agree, but a lot of people see Amazon as sort of the gold standard here. Other people don't like the experience. I happen to like it myself, but what have you learned from sort of how the Amazon takes the approach? You know, what can you borrow from that metaphor and what can you improve upon as an industry, not even necessarily specifically guests? Well, I think we definitely, I think as retailers, we look up to what Amazon has been able to do, not only in their analytics, but also the customer experience and what, you know, they really drive customers to want to come back and shop with them. And they leverage technology, I think, in the best ways. So definitely for us, it's to look and see how we can incorporate some of that into our business. Bruce, you mentioned, I'll say knowing your customer, it's funny, you know, with all this big data and social data, Dave and I were working for a couple of years on a project with Twitter monitoring, crowds and crowdshats and crowdspots. And, you know, for the first time in modern business, you can actually monitor people and businesses end to end. And most people don't really know who their customers are. They've had studies and blind studies and panels and kind of the magic science. What have you learned about big data around understanding the customer and how to serve end users? That's different from when you entered into the industry. I mean, you're in the middle of the transition on the beginning of this major reflection point that's going to change how business is organized with big data and all the different architectures, changes that are going on. What's your personal view on, you know, when you entered in the industry as a fresh college grad to now an issue leading executive? I mean, what's different in the tech and the approaches or mindset tech, business model, I mean, all three, what do you see? What's your personal view on that? Well, really, I think at the core we're still trying to solve the same problems, right? How do we make data actionable? And it's just, we're given a different set of tools, a different set of data, but actually I think now with the convergence of all this technology and all this data, we're able to actually do something with it. In the past, we'd always have to, you know, create small samples of data and, you know, we were limited by technology. Now it's actually very refreshing because you can actually do something with this data and to try to really understand what our customers are doing, at the end of the day, you know, in a way, people get scared from that. You know, it's like, that's a lot of data or, you know, how do you use social data? I think for us, it's like we have to make sure that, you know, at the end of the day, we have the customer in mind. Like it has to be customer-centric and we have to be very careful with that data, you know, and earn their trust. So what if you could break it down for a lay people in the audience? You mentioned, you know, you migrated from a traditional database architecture to Vertica. So why Vertica? What was compelling about that, you know, from a technology or architectural perspective? What, as a practitioner, appealed to you? Well, first of all, we had a lot of performance pains with our Oracle system and we thought, you know what, we can't really grow the business if we keep having these performance issues, our staff is dedicated to performance tuning all the time. Our end users, our, you know, our analysts, they couldn't ask the right questions with the systems that we had because it was so slow. So really, when we looked at migrating from Oracle, we looked at a couple of other databases as well, but we settled on Vertica because we felt that they had a very, very clean approach to the technologies. Like if you wanted to query data fast, their approach just seemed just a little bit smarter. And I mean, they had, you know, Mike Stonebreaker as one of the founders, it's, I mean, that says something, right? I mean, he's up there in the industry. As a CUBE alum, we've had him on. And so what do you mean by clean it from a practitioner's perspective? How do you sort of evaluate, you know, cleanliness, if you will, in a database architecture? It just seemed, you know, when we looked at, we looked at other databases, but the speed at which you could query and just the way they compressed the data, the very little tuning that you actually have to do. I mean, there are other claims when we first started to do the PLC. We were very skeptical as well. You know, in fact, our DBA who's at this event, he was actually extremely against us looking at Vertica. Don't even bother. Oracle's the best, you know, we can just keep tuning it. But now he's actually here and he, you know, if you ask him, he loves Vertica. He thinks it's amazing what this architecture and how it's made. And it's just, it's a very smart database and it's a very smart platform, how they engineered it so that, you know, the amount of disk space that we're using, the performance that we get out of it is just huge. And they have, you know, a stack of analytical functions now that we're leveraging as well. So our staff doesn't even have to, you know, be, it's easy for our staff to really run, run queries with all these analytics. Yeah, one of my questions is, is it the nature of the data that you're running? Or is this generalizable for many, many, many use cases? And obviously many use cases, but many general purpose use cases, which it sounds like yours is pretty general purpose. I think so. I think we, I mean, I think we have the same problems that probably 80% of the retailers have out there. I don't think our data is that much special, more special than other people. So I really think that it's just able to hit the sweet spot in terms of performance and also flexibility in terms of being able to provide these answers to our users much more quickly. And, you know, the architecture, we've installed Vertica and a couple of other countries as well. In fact, we're going live with Korea in a couple of weeks. So it's been really exciting to see how, you know, the other teams are also around the world for gas are able to use it. So you mentioned, just mentioned Korea, you mentioned Japan earlier. How are the cultural aspects, how do the cultural aspects of data play into your strategies in terms of how you get adoption and how you accommodate the business needs? Is there a culture that must be, but can you talk about that? Yeah, there are actually, it is quite, it can be quite challenging. We spent a lot of time working with our, the teams in the other countries and getting them to adopt data. Different countries have different propensities and different levels of trust with running queries and getting that data versus looking at their spreadsheets. So getting them to adopt also the master data to, you know, look at the same calendar to, you know, like the retail week ends on a Saturday here. Well, it's different around the world. You know, they don't want to end it on Saturday. They want to end it on a Sunday. You know, so just those little nuances. I mean, it drives you crazy when you're trying to implement. But then as soon as you show them what you can do with this technology and they begin to see, wow, I can save so much time. They start to really, really, the adoption is much easier. But there are a lot of cultural nuances. So take that nuance that you just mentioned, you know, closing on a Saturday versus a Sunday. I mean, obviously it requires, there's a development, you know, effort required there. What's your philosophy on small teams versus big teams in the development environment? I mean, I know the rage is small teams, but I wonder if you could talk about that a little bit and what experiences you've had in that regard. It's a good question. You know, we actually do believe in small teams and small local teams, because only with the small local teams can they be agile enough to add the requirements that their business users need. For us, it's, you know, we went away from a centralized approach. So we have small local teams with, you know, our database hardware and our BI software out there as well. So it actually makes sense for what we're doing. We do work with them a lot so that we share a common data model so we can extract the data back for our corporate purposes. I know we're tight on time, but I wanted to ask you a question. You know, John and I often ask practitioners like yourself, sometimes we call you tech athletes. We've got a lot of young people in the audience. We're always encouraging, you know, obviously everybody's pushing math and science, but what advice would you give to young people looking to get into the data field? What advice? Take some sequel classes, get an internship, you know, get as much experience as you can, because I think a lot of it is also business experience. You know, so people that come out straight out of college with no work experience, no concept of an office or business or how businesses make money, I think they're at a disadvantage. They have to know both the business side and the data side to really be successful and they get our field. How about, same question for fellow practitioners. People, you know, that maybe organizationally have, you know, folks that want to hang on to their legacy systems. What advice would you give to somebody who's sort of an innovator like yourself that wants to, you know, move their organization forward, but is cognizant of some of the tensions and arrows that they might take? You know, it's never easy. I've talked to a few other, some quick service retail food companies that actually, they still use cubes and, you know, they need to just take that step, you know, because no one's going to be really, I mean, with the technology that we have, they just need to make that step and it wasn't easy for us when we did it, but, you know, we're so much better off because of it. All right, Bruce, thanks very much for coming on theCUBE, really appreciate it. Thanks for really stopping in, John. Dave and I really appreciate your time and good luck with everything going forward on your big day. This is theCUBE. We'll be right back with our next guest after this short break. Great, thank you.