 Live here at .conf Splunk's user conference, we're at the Cosmopolitan Hotel in Las Vegas. I'm here with my co-host, Jeff Frick, I'm Jeff Kelly from wikibon.org and we're really happy right now to welcome CEO of Splunk, CEO and Chairman, Godfrey Sullivan. Thank you, glad to be here. Well thanks for joining us, really appreciate it. So geez, what an event, 1,000 plus people, you guys are doing quite well, you had a great quarter just ended after a successful IPO in April. So how do you feel, you guys are on quite a streak right now? Well the thing that always makes us the happiest is when we have a lot of customers around us. So having 1,000 customers here in Las Vegas has everybody in the Splunk world just thrilled. So we've added 400 new customers in our most recently completed quarter, so 4,500 customers is just, that makes us really happy. Yeah, that's great. And also the fact that you've got repeat customers, because as an old time salesman you know it's always easier to get the repeat order than the first order. So the fact that you're landing in at a low friction point or a relatively easy high value solution and then seeing your customers, I read in the analyst call, continue to expand the use of the platform. Because it's something they couldn't do before or they just see the bright light pops on that wow we can do this and we can do that, that's got to make you really happy. Well Splunk's a really unique product in my experience. Where else can you download something off the web, use it for a while, installation is a few minutes and going live in a small instance is maybe a day and then grow with that experience over time. And so a lot of our customers who convert from a free license to some sort of paid license typically only spend $10, $15,000 or something like that. But you heard in our quarterly earnings report that 70% of our revenues came from our existing customer base. So it's pretty amazing that a product can be that easy to download, that quick to install and get running and yet scale to the kind of volumes where our customers are learning from it and then expanding their use of it and of course they pay us a little bit when that happens but it's a mutual benefit relationship and so yeah we're very fortunate. So in your keynote this morning you said something which I found interesting that three years ago trying to explain to a CEO at a company what machine data is, why it's important, how you can leverage it for value. It was a very kind of a tough conversation to have and it's gotten a lot easier over the last three years. Can you talk a little bit about why that is and explain to our audience what really makes the machine data specifically such a valuable asset to exploit? Sure. It's been a challenge to say to someone you should look at your log files. Let's just take machine data down to its slogan or it's the slang terminology for a moment. If you said gee there's to a CIO there's a lot of data and value in your log files and they just say really? I mean I was back when I used to run Hyperion we were a software development company we just thought about log files as being those things that the developers spit out to try to find errors in their code or some sort of other and as long as they were happy we were happy but you never thought about that being a we were a BI company we never thought about getting intelligence out of logs thought never occurred to us. We looked at a lot of data and we delivered intelligence out of a lot of data all of it on the structured side you're looking at customer names and product ID codes and comparing revenues from one quarter to another you never think about looking into a log file to see if there was something of value in there. I think what Eric and Rob and the founding team really did that was so innovative was to say if a transaction went inside one side of the IT blender and never came out the other side why not and they started to use that as the foundation problem to go to go create the search engine and from that found that there was a lot of value in that data so it's a kind of a new concept and and as a result it's taken a while to educate the world on gee there's real value in your logs there's real value in your in your IT data or your machine data your device generated data and that's just exploded now with the internet coming of age internet transactions mobile devices that the amount of volume that has is now being created has just exploded but now execs even old software execs are beginning to understand there's a lot of value in that machine data if you just harvest it analyze it and and ask it a question right right so you're doing a great job in taking advantage of the data that you're before can cannot be used right what's the next big challenge so we're having got a great infrastructure people are buying the software they're using the software it's innovating like crazy so from your point of view from the CEO point of view what's the next hill kind of what's the next not necessarily over the over the horizon hill but the next kind of challenge that you and the team are trying to to take down sure well what's becoming mainstream is that the IT organizations around the world are starting to understand what's the plunk is you know the IPO is really helpful to us in terms of it's a branding event it raises your profile of the brand itself the name of the company so we've really seen an increased awareness an increased recognition factor folks are saying oh I heard about you I read about you that that's really one of the things that an IPO does for you besides give you a new set of investors is actually is an important branding event so that's really helped increase the velocity of our mainstream business think of it as the IT version of machine data right sort of over the next hill we talked about in the keynote this morning which is what about all those uses beyond IT where other types of machine generated data could provide real value like the heart the exhaust of a jet airplane engine or the information coming from your car's computer or the information coming from a medical device information coming from from other types of mobile devices that so our social media streams as we have customers who are actually presenting sessions here on things like building management you know what what's the operational intelligence that comes out of an elevator right so all the devices of the world the machines are starting to spit out digital data and that sort of the next thing is how do we both do a great job with MIT but expands plunks vision and reach beyond IT to enable all the rest of the machines of the world to be able to talk back to us and have us listen to them and understand where their value is and just thought we're all kind of virtual machines now because we're carrying around the mobile devices we're logged in we're interacting with applications and things all the time so then like I said the machine is no longer just a machine it's the connectivity to us that's generating so much and we can't live without it so I'm interested to talk a little bit about so where's some of the implications for the enterprise we're talking about all this different types of data in different use cases it can be a little overwhelming so what are some of the challenges in terms of for your customers in terms of maybe culturally or even organizationally putting themselves in position where they can really start taking advantage of this kind of new capabilities that Splunk provides right it's really a learning experience you know it's it's like when the one of our customers came in from from Japan and I saw her in the hallway and asked her what why are you here why are you in California and she said well we've Splunk has made us aware that we need to listen to our machine data and so we think we can take the elevator data that's kind of our elevator data and create intelligence out of it I said how in the world do you do that she said well you know everybody has a car a personal access card and the elevators all have computer controls in them now so all we have to do is Splunk all that data and we'd be able to see which buildings are in most danger which buildings have economic either energy coming their way or risk we can look at traffic going to floors or into specific suites gives us a better leading economic indicator of a building's health in fact she said the elevator is probably the best way to understand the future economic health of a building you think wow that's a pretty good concept I never thought about elevators that way so our customers continue to give us all these new exciting ideas about what where the devices of the world can help us so you know but the big challenge for customers is just thinking about it that way right how do you learn about this new type of data and how do you then think about how to get the intelligence out of it it seems very kind of chicken and egg and you know how did she think of that you know who was it some some guy that works at the at the at the lobby that noticed that suddenly no one's coming in and out anymore the building and sure enough the company defaults in the building comes available or is it is it now being able to play with the data if you will in ways that you couldn't before that then provides the insight to think of new ways to do it you have any idea and the elevator case kind of who came up with that idea and then actually was able to validate it with software well I don't know in that case but I'd say in most cases what our customers it's that we start an IT the IT guys get to know what it is and then the light bulb goes okay they go wow if we can do that with Splunk then we need to go talk to our colleagues and other departments right and so Splunk is a good example of a unique technology that kind of tears down the organizational boundaries so before Splunk the guys in security had software that was narrowly used for that purpose and they rarely had a reason to go down the hall and talk to one of their peers in another department about it except just to update a status or something Splunk is one of those rare technologies where the more data you put in the more value you get back so now the security guys have a reason to go down and talk to the people in applications development or in infrastructure or maybe even in this case over into operations and say hey if you guys thought about your data this way so I'd say in most of our customer cases it starts in IT the guys over there that light bulb goes on they get it and then they start reaching out to other departments to say hey have you guys thought about this so it's pretty cool so you know one of the things that we're seeing is definitely a lot more interest from the enterprise in terms of virtualization we were at VMware not too long ago I think actually Splunk was there give a presentation as well that's right but I'm interested you know so Splunk kind of got it start back in the mid-2000s but now you know virtualization has come on strong over the last couple of years so how does that impact your what you do in terms of evaluating an infrastructure IT infrastructure does that add complications and how does Splunk kind of adapts in this new kind of world of virtualization cloud computing well virtualization adds one more layer of complexity to the IT stack so while you're getting the economies of knowing that you can run multiple VMs inside of a physical device you've now abstracted away from the physical device some of that information so when something goes wrong and you're with your application or your security or your customer accounts or any other type of activity how do you how do you get at that data so we've worked closely with VMware and Citrix and other companies in the virtualization category to say help us get that data so that we can then ingest it analyze it correlate it with other other types of data so to us it's just another very important data source another one is of course big data kind of big data analyzing big data I know you've got a lot of partners here and other firms you work with Hortonworks among others are here at the Hadoop world how does a kind of Splunk kind of fit into that larger big data landscape and you know how what is your relationship with kind of the Hadoop side of the equation do you work with helping you know Hadoop administrators kind of understand the Hadoop environment which really is one of the big issues in that particular field in terms of adoption and spurring adoptions kind of the management tools and kind of understanding what's going on inside your cluster sure well I'm on the road all the time and talked to lots of IT execs and one of the most common things I hear is we have lots of data we don't know what to do with it so we're standing up a Hadoop stack our Hadoop cluster and we're going to put a lot of data in there and if we ever need it or if we have to go back and ask it a question at least we'll have it if you ask them well what was there a specific problem you were trying to solve often they haven't gotten that far yet they're just knowing they need to fire hose in lots and lots of data and so it so Hadoop is a beautiful cheap batch storage system it's a great file system for just putting all kinds of different data into it and it doesn't cost anything from a software standpoint to do it costs a lot from a hardware standpoint development and how to get it back out so we think this is a huge opportunity to provide value to those those big data sets which is to connect Splunk to it so that you can from a really easy user interface you don't have to hire lots of data scientists or do lots of programming work you hook Splunk to it you say go get that data let me analyze it the Splunk way or maybe I've indexed it through Splunk on the way in and then you go drop it into the Hadoop cluster for long-term cheap storage so we think there's a big opportunity to both add value to that universe by being able to analyze that make it easier to analyze that data and also be able to monitor those big stacks you get those big clusters in place and they and they break I mean things go wrong it's a complex application file environment right so we might as well apply Splunk to it that's what we do right so we look at it as a very interesting complementary opportunity good I want to shift gears a little bit you mentioned it a little bit in the keynote on Splunk for good and and obviously as a successful company with a few dollars in the bank and it sounds like a really great culture you have an opportunity that you can look outside of kind of your core customers and your core business activities and and and obviously somebody said global warming is a big data problem I heard that quote the other day so can you talk a little bit about Splunk for good and some of those efforts that you and the company are doing to take advantage of what you've created love to well Splunk for good simply put is our ability our our hope our aspiration to be an advocate for making data open as often as possible and being able to reach out to people like state governments local governments to be able to publish that information so that's accessible to the community at large the more data that's published the more knowledge and richness there is available for all of us as citizens and so for example we have a I had a customer a partner in from Rio just last week and they were working with the local government down there to understand road sensor data traffic signals and everything else it's it's all there it's all available if the if it will be published once it's published we can index it into Splunk and they are they are contemplating using Splunk so that they can better manage traffic so just think about how bad traffic is in Rio and and parts like that just any major metropolitan area if you're actually going to have an impact on global warming or just traffic density or people's ability to get to work you need to understand when a traffic signal fails or a road is blocked or some other thing how do you how do you get that information what you do with it how do you take an act proactive management action towards it so you can remedy that situation we can provide an enormous amount of value there and we have to make sure that especially governments are get you know start to be more and more open about publishing their data the other thing we're doing with Splunk for good is we're advocating it to nonprofits all over the world if you need Splunk here's what we can do for you come get free free software and we'll help you get it get it up and running and our customers are involved at the community level with nonprofits everywhere so for them to know that and know that they can do some outreach there would be really good so we're we're excited about it's great so you know in terms of looking forward obviously we're in kind of a tough economic situation here in the US and Europe's troubles are well known maybe worse yeah could be a little bit worse so but you know I mentioned in our intro earlier today I believe at 19 straight quarters of double-digit revenue growth and Splunk so that's pretty remarkable that goes right through the the recession and the downturn so what's your kind of outlook going forward in terms of you know business especially when we're seeing some of the larger kind of what you might call mega vendors IT mega vendors are seeing a pullback in some spend how does that impacting your business and big data in general I don't want to go too far and say big data is recession proof but is it big data isn't recession proof customers have to have a need for something specific or they won't spend money on it so I don't see I don't see big data projects per se being recession proof as we look out though we have a very ambitious product product agenda so we have teams building Splunk enterprise we announced Splunk storm a few weeks ago for general availability we're expanding that team that's your cloud based our cloud offering our SAS offering we're building apps and content and we have a new team that we're investing in up in Seattle to build the SDKs to help extend Splunk further and for developers so with we have a very ambitious product agenda and customers appreciate that and are asking us for more we can only do that by investing really heavily so part of what we've told our new investors from Wall Street is you'll have to bear with us because right now we're in a high growth phase and we don't expect to be profitable will be cash flow profitable but not gap or non-gap profitable we are investing every dollar we make into additional development resources field coverage professional services support so when you're when you're expanding your customer base as quickly as we are you have to stay in high investment mode so you know none of those businesses are not no businesses recession proof but we happen to be in a high growth mode and we need to continue to invest as heavily as we can to match the customer requirements so all right well thank you so much I think we're just about out of time thanks so much for joining us today really appreciate you know you're very busy here at squunk.com now here comes the fun part which is for the next three days we get to go attend those customer sessions and find out all the creative things our customers are doing with Splunk and that's the most fun part about it for those of us that work at Splunk we're going right to the customer sessions you know when they're done with their sessions then I'm right over the queue we'll blast it out to our community so thanks so much thanks really appreciate it and we'll be back after just a moment with our next guest