 Okay, we're back here live in Las Vegas for SplunkConference.Conference 2013. This is theCUBE, our flagship program. We go out to the events, and extract the student from the noise. I'm John Furrier, the founder of SiliconANG, I'm Joe, my co-host. I'm Dave Vellante, hi everybody, wikibond.org. Clint Sharp is here as the Director of Product Management at Splunk. Clint, John's been busting my chops all week about calling me hunk. You are the real hunk, so congratulations on having a great conference, and welcome to theCUBE. Thank you. Thank you. So how do you feel? You saw the keynote yesterday, of course. Oh, it's fantastic. A lot of buzz up there. Oh, absolutely. It was, the energy in the room was fantastic. I love our customers. They're so excited when they come to users' conference, and you can hear the enthusiasm in the room of the keynote, and I've been walking around talking to customers, you know, constantly for the last two days, all very interested in our Hadoop story, and helping them analyze the data they've already got at rest. Well, what I've been hearing from customers, I mean, when we ask them what's their big takeaway, the one thing that really comes across is I could be using Splunk for a lot more things than I am today, and so talk about how hunk fits into that. Yeah, sure, so at Splunk, we've built the world's best machine data processing engine, and many customers get started with log search, and maybe they stop there, maybe they progress further, but they think of it as a tool that they use to work on their applications and IT infrastructure. We've had customers for years doing all of these types of use cases, but it takes a little bit of prodding to kind of get them up the curve, and what we've done in the last keynote, I think, has really opened their eyes to, wow, I could be doing business analytics, operational intelligence on top of my data sets, because now that I think about it, it's very easy for me to take that same data and make it relevant for the business. Well, I think what's happening, too, is a lot of customers, they'll bring it in, so there's some champion, maybe he or she will fly in under the radar, not even have to do a justification, I just need this to make stuff better, and then they'll be off in the corner doing their thing, and all of a sudden somebody else sees it and says, oh, what's that, what's that? And then it starts to expand, and you see that today with a lot of successful companies, smaller companies, but growing, guys like you, you see the similar dynamic in Tableau, ServiceNow, you guys got it right, you got a 10x value proposition and they're exploding from there. So, what gives you confidence that now you're out, the cat's out of the bag, everybody sees, all right, Splunk, real company, doing all this business, growing to 50% a year, now everybody's saying, oh, we can do that, too. What gives you confidence that you can stay ahead of the competition? So, yeah, everybody said, okay, so I'm going to glue a new big data back in on my product and say, I'm the next Splunk, they're all going to say that and that's great, but I'll tell you that the secret for what makes Splunk so successful is the way we look at data. We think about data very differently than the way everybody else thinks of data and I don't mean just schema less, no SQL, et cetera, even in the no SQL stores, I'm still defining a structure and I'm putting data in a semi or poly structured form. Splunk is reading that at search time, I'm building my structure at search time, it's fundamentally very different and you see when we talked about Hunk at the keynote, we talked about exploration. What makes that so unique is we go and search for the data, we bring it back, we structure it then and nobody else is really doing it in exactly the same way. Many people are doing schema on read but you still have to have that schema first and so it's a very different way to look at data and I think that keeps us just way out ahead because users just love the search bar experience, they love to be able to just look at their data that way. So you're essentially building that in on the fly, allowing the flexibility so that you're not locked into a schema? Absolutely, I mean if you look at, let's talk about business analytics in that marketplace for a few minutes. When you look at a typical BI project, 60, 70% of the effort in a BI project goes into building rows and columns and structure and trying to decide what questions you want to ask in advance. If I build a data warehouse, I have to define all my tables, start schemas, get everything, build my aggregates, huge amount of time and effort spent on ETL, spent on designing data and our approach is different. We say dump the data in whatever format it came as dirty, as nasty as you can possibly hand it to us or structured, we work with structured data too. Dump it somewhere we can get access to and then let's work to build your reports and then we can take you all the way to structured analytics. But let's start there and let's take that money that we've invested in ETL and tables and schemas and throw it into people analyzing data. And so the best practice there is iteration, right? Talk about that a little bit. Yeah, so a lot of what we heard when we went into starting to look at the Hadoop marketplace was, we've been successful getting data in but now what I need the ability to do is search and explore my data because now I've been successful in getting the data in and I've been so successful I don't know what it is now. So we wanted to give users the ability to go explore that data set and iterate upon it as they discovered what they're looking for and as opposed to saying I want to select splat from this table and I build all my different tables, that means I must know what data is there to create a table on. And so we thought that the much better way of going about this is just giving them a search bar. I just want to find the data that I'm looking for and then I'll work to structure it, then I'll work to do analytics on it, create dashboards, pivot on the data, et cetera. Talk about the relationship with the cloud product and the importance of cloud, as the old saying, big data gives cloud something to do. How important is cloud to you all? I mean the basic value proposition or one of the concepts that you guys are putting forth is leave the data where it is, we'll lay hunk on top of the Hadoop cluster and you don't have to move it. So where does cloud fit? So cloud is a key part of our strategy and I think what makes us, a lot of our competitors are going to come after us and say, hey, we're true SaaS, we're a SaaS play, we're the only company in the business doing what we're doing on premise and in the cloud. And what we're hearing from customers is, not only do I want to have my data in the cloud but I want to have my data on-prem, some data's on-prem, some data's in the cloud, I want the data to stay where it's at, we're going to give you the ability to search across both of those through one seamless experience which nobody else in the marketplace can offer. Awesome, so what else are you hearing from customers? You're in beta, right now, so you've gotten some initial feedback, what are they telling you? They're telling me that they want us to do more. They always want more, those customers, they're always wanting more. We're going to work on- That's good, when they tell you they don't want anything else, that's when you got to run. I'll tell you the most frequent request I'm getting right now is, okay, you've done it for Hadoop, what's next? Where else can you go get my data? I want data in Cassandra, I want data in Mongo, I want data in HBase and Acumulo, so by far that's where they're looking for us to go and expect us to look pretty strongly in that area because the way we architected Hunk and our virtual index technology is completely modular, I can easily substitute in differing backing data stores which is incredibly unique for us in this marketplace because everyone else who's going and doing pure play Hadoop analytics the Hadoop is very baked into what they're doing and for us it's not, it's one backing data store that we're looking at and expect us to look at other places to go get the data. So, Clint, talk about the roadmap, obviously you mentioned people are demanding more features, obviously Splunk's got good success there. Visualization is always a hot topic and analytics is a killer app, you guys have got a great position, people just dump their data and they Splunk it as they say, it's a verb now, so you got good traction there, but in front of the product management side, how do you prioritize the roadmap? I mean, you've got Visualization, it's still hot, people want things more visually, more laid out, they want better analytics and then you've got other platforms, data platforms to support. What's that balancing act there? It's a tough balancing act because obviously most of our customers who are here have come from an IT background, they bought us for IT search, they want us to continue to invest in application management and operational intelligence, so we're investing some into our analytics play because we're there, we're still keeping in mind and my background is running IT, before I was a product manager, I ran IT for a telco and so I'm always keeping that IT guy in mind, so we're trying to make sure that we're designing an experience that's horizontal and benefits everybody at the same time, it's a really difficult balancing act, but we're continuing to come out with great new content as well, with VMware, our new Unix app, I mean these are things that are very central and core to the IT experience and so we're splitting it as best we can. Talk about the mobile, obviously bug sense, big news, we had those guys on, really energy was just amazing, mobile expertise is critical as you build that into the roadmap, how is that going to morph into the plan? Yeah, so well, mobile's everywhere, right? And not only is it everywhere, but it's generating a ton of data and as Splunk we want to make everybody's data valuable and accessible, so we're taking our concept of forwarders that we use inside of the agent and we're going to go work to get data out of the mobile platform because we're already helping them gather data on mobile applications on the server side and what our customers are telling us is that, if I'm gathering the data on the server side, that's great, but I need to be able to correlate and that's something that Splunk does very, very well, I need to be able to correlate what's happening on the actual client with what's happening on the server so that I know that if my customers are loading up a bunch of bad software on their iPhone or Android device and that's causing me application problems, I want to make sure that I can prove that definitively versus having to say, well, on the server side it all looks good, right? So we're going to keep investing and getting detailed data off of the mobile device. We had a chance last night to sit down with some of the folks in the product team at the party and get to know some of the folks and we commented yesterday, we wrote a blog post on SiliconANGLE about the product culture at Splunk. It's very product centric, excellent, it's a mindset and it shows and you hear people kind of celebrating these ship six, you know, take a little breather, let some steam out, kind of like college final exams, kind of take a little break. But I got to ask you, going forward now, now that that's behind everybody, what's the guiding principle on the product teams now? What is the Splunk internal, you know, in it to win it? What's the slogan? What's the guiding principles that you guys put forth to your teams? We're a quirky culture. So we're out there, I don't know that we have a slogan, we definitely have a mascot. So, and we've got a new hunk mascot as well. I don't know if you saw the slide at the end of the presentation. So we're, was that your body? Unfortunately not, unfortunately not. But we're going to make up some bobble heads and we're going to put them on our desks and we're going to go keep trying to win in that marketplace. So I think it's, you know, work hard, play hard, I would say, and that sounds cliche, but when I, you know, I leave the office at a 6.30, 7.30 at night and the developers are there until, if you come back at midnight, there's people still there coding, right? And so, we love the product ourselves and that's what I think is so unique about Splunk and our products organization is we use the product, we love the product and it's a, it really shows in what it is that we're building. Well, do you think, so let's talk about the market. So, you know, we're always trying to squint through the market sizing and what the TAM is. How does the hunk capability affect that TAM? So obviously we talked about the Hadoop piece, but you're defining big data as more than Hadoop. So how do you think about the total available market and how it's affected by hunk? Yeah, so I mean, I honestly, over the next 10, if we go on a 10 year to 15 year horizon. Yeah, let's keep it far enough off so that nobody will really remember what we said here in theCUBE. If we push it out that far, then potentially all data is accessible to hunk and our ability to go get it, right? So, but we're really focusing on a little bit, we don't want to become the next BI tool. If you look at other companies that are doing BI, they're doing a great job and it's also a really, really heavy market. There's a lot of competitors in that market. That's why we articulate our message around operational intelligence and being able to look at the data that's relevant to my business right now. And hunk's expanding that a little bit to say, let's look at that operational data over the course of a year, two years, five years, as opposed to trying to build structured data marts and that sort of stuff. We're really hoping that they spend more time just analyzing the data. I think potentially over 10 years then all data is potentially open to us. So let's unpack that term, that phrase operational intelligence. What do you mean by that? What's underneath all that? So we're going to invest more in the idea of operational intelligence as a product as well as just an idea. And when I say operational intelligence, I mean I want to know right now how my business is performing. So when I used to work at my last job, I ran application operations. I could tell you down to the minute what retail store people were buying things at, what calls to care were happening, from what markets. And the end all be all of being able to look at the health of my businesses. You know I may have speeds and feeds monitoring, I may be looking, I may have a whole suite of tools looking at my IT infrastructure, but if widgets aren't walking out the door, then I'm failing, right? And that's sort of the meeting point between the business and IT. IT is running the systems that feed business processes and the business is looking at those business processes. We're going to help them meet in the middle. So you're talking about instrumenting the business all the way through. Looking at detailed data, and we have customers doing this today all over the world looking at business processes so they can see exactly how things are functioning right now. So the business is looking at it from a real-time business analytics perspective. The IT guys are looking at it from a perspective of now I know for sure whether my systems are actually functioning because ultimately I don't care if my API is responding at a 100% success rate, if my sales are down 30%, something's wrong and I need to go look. So you're saying you've got examples today where you can instrument from the IT infrastructure through the applications to the business process out to the customer and have essentially a value flow, if you will. Absolutely numerous examples of that. In fact, I know you can do it because I built it. So give us some examples of how people are doing that. So we have a large telco customer in another continent that's looking at all of their order to activation process, right? So it's really critical if I buy a cell phone that I walk out with a working phone. We have customers from a large healthcare insurance provider who's using us to look at claims processing and watching through each step of the claims process, are claims falling out? Are we processing orders in a timely fashion? And it was visibility that would have taken them years to build inside of a BI platform, simply because the data is coming to them raw and unstructured and they need to be able to look at it in the way in which we excel at. Is there an anticipatory aspect of that as well where an event occurs, let's say somebody buys a cell phone and your customer is able to anticipate what the next step is going to be. So when I call up, they know what I'm trying to do. Is that, are we at that point of operational intelligence? We're not there yet, but we want to go there. So one of the things we really want to do in terms of enabling our, and we can do a lot of this today, it's just a little bit more esoteric than we would like it to be, is in terms of throw a bunch of data at me and tell me what's weird, right? Just show me, if I have a trend line that's going this way and I'm two standard deviations plus or minus, the system should just tell you, right? And so we have a lot of those capabilities in our search language today, you just have to go find them and so we're going to work to start making that easier for customers to get access to. And then ultimately automate that. And so yeah, exactly. And throw me an alert anytime something weird happens and we should be able to do that because we know what, heuristically we know what the norms are and we have those capabilities today and now we just need to bubble them up better to the end user. Okay, Clay, we're getting pushed on time. Final question to end the segment for you is what is the number one priority right now in the product crew? Obviously in terms, it could be feature, it could be polishing up simplicity, analytics. What's the number one thing you guys are focused on right now? So for the next release, what we're focusing on is end user delight, right? We've paid off a lot of technology debt with this release. We've got a new UI, a new framework. We're able to move very fast. We've built a great new business user interface. And so what we want to do is we want to get feedback from the people who are using the product and we want to, we just want to delight them, right? When they open up the next box that we ship them, we want them to say, man, this is just much better, better forward management, better clustering. You know, the things that matter, you know, anomaly detection, things that are, things that are just going to make their lives easier because we know the IT market and we know the analytics market and we know what makes people happy. And we just, we want them to, we want them to smile the next time we give them a version. Well, congratulations. It's certainly exciting for the queue because it makes our job really easy when you've got customers that sit next to us and tell us how delighted they are. And they say things like, Splunk allows me to dream. And that's the kind of sentiment you guys are enabling. So congratulations on the product's success and congratulations. This is theCUBE Live Splunk Conference in Las Vegas, date two of two days of coverage, exclusively on SiliconANGLE.tv. We'll be right back. This is theCUBE.