 Live from Orlando, Florida. It's theCUBE. Covering.conf18, brought to you by Splunk. Welcome back to theCUBE's coverage of Splunk Conf 18. It's Florida week. I'm Stu Miniman and my co-host for this week is Dave Vellante. Dave, I'm really excited. You've done the show a handful of times. It's our seventh year doing theCUBE here. It was, it is my first time here. I thought I understood a few of the pieces of what's going on, but it's really been crystallizing to me. When we talk about on theCUBE, for the last couple of years, data is at the center of everything. And in the keynote this morning, they talked about Splunkers are at the crossroads of data. I've talked to a bunch of practitioners here. People come to them to try to get access to data. And the vision that they've laid out this week for Splunk Next is how they can do a massive ham expansion, try to get from the 16,000 users that they have today to 10x more. So what's your take bin on where we are today and what Splunk of the future looks like? Well, so Stu, as you know, the keynotes are offsite about a half hour away from the hotel where we're broadcasting. So then there's like 8,000 buses that they're jamming customers in. It's a bit of a pain to get there. So logistically, it's not ideal. And so I thought the keynotes today just remotely, we didn't hop in the bus because we had to miss a lot of the keynotes yesterday to get back here. So we watched remotely today. It just felt like there wasn't as much energy in the room. And I think that's for a couple of reasons and I'll get into that. But before I do, you're right, I've been, this is my fourth .conf. And I was struck in the audience at how few people, actually it was probably less than a third of the audience when they asked people to stand up had been to four or more .confs. Ton of people, first year, second year. So why is that relevant? It's relevant because these are new peoples. The core of Splunk's audience are security people and IT operations management people. And so with that many newbies, newbies, they're trying to learn about how they can get more value out of the tool. Today's announcements were all about line of business and industrial IoT. And frankly, I think a lot of the people in the audience didn't directly care. And I'll explain why it's important and why they actually do care and will care going forward. But the most important thing here is that we are witnessing a massive TAM expansion, total available market expansion for Splunk. Splunk's a $1.6, $1.7 billion company. They're going to blow through $2 billion. This is a playbook that we've seen before out of the likes of, particularly ServiceNow. I'm struck by the way in which Splunk is providing innovation for non-IT people. It's exactly the playbook that ServiceNow has used and it works beautifully and we'll get into some of that. So Dave, one of the things that really struck me, we had seven customers on the program yesterday and the relationship between Splunk and the customers is a little different. You always hear, oh, well, I love this technology. Lots of companies, you've been telling me how passionate we're, but really partnerships that you talk about, when you talked about, we had an insurance company from Toronto and how they're thinking about how the security and risk they look at, how that passes onto their customers. So many, it's not just people are using Splunk, but it's how it affects their business, how it affects their ultimate end users and that value of data is something that we come back to again and again. So the classic Splunk user is somebody in IT, IT operations management or the security knock. They're hardcore data people, they're looking at screens all day and they love taking a bath in data and Splunk has completely changed their lives because rather than having to manually go through log files, Splunk has helped them organize that sort of messy data, as Doug Merritt said yesterday. Today, the whole conversation was about expanding into line of business and industrial IoT. These are process engineers. There weren't a lot of process engineers in the audience today. That's why I think a lot of people were excited about it. I'm super excited about it because this is going to power, I've always been a bull on Splunk. This is going to power the next wave of growth at Splunk. Splunk is a company that got to the public markets without having to raise a ton of capital. Unlike what you're seeing today, you're seeing hundreds of millions of dollars raised before these companies IPO. So Splunk today in the keynotes, first of all, they had a lot of fun. I was laughing my, you know, what off at the auditions. I mean, I don't really, some of that stuff's kind of snarky, but I thought it was hilarious. What they did is they said, well, Doug Merritt wasn't a shoe in to keynote at this. So we auditioned a bunch of people, so they came in and people were singing, they were goofing, you know, hello Las Vegas, we're not in Las Vegas, we're in Orlando this year. I thought it was really, really funny and well done. You know, Stu, we see a lot of this stuff. And yeah, absolutely. Fun is definitely part of the culture here at Splunk. You know, love that, you know, that we talked about yesterday, you know, the geeky t-shirts with all the jokes on that and everything. Absolutely so much going on. But, you know, Dave, there's something I knew coming in and we've definitely heard it today in the keynotes. Developers are such an audience that everybody's trying to go after. And you talk about kind of the traditional IT and security might not really be the developer audience, but absolutely that's where Splunk is pushing towards. They announced the beta of the Splunk developer cloud, you know, a number of other, you know, products that they put in beta or are announcing. You know, what's your take as to how they go beyond kind of the traditional Splunk user? Yeah, so as I was saying, I mean, this is to me a classic case where we saw this with ServiceNow who's powering their way through five billion land and expand something that Kristen Shabo, former CEO of Tableau, used to talk about where you come in and you get a foot in the door and then it just spreads. You get in like a tick and then it spreads to other parts of the business. So let's go through some of the announcements. Splunknext, they built on top of that today. Splunk business flow. They showed what I thought was an awesome demo. They had a business person, you know, it was an artificial example of the game company. What was the name of the game company? Yeah, Butter Company. Butter Company. So they took a bunch of data, they ingested a bunch of data on the business workflow and it was just that. It was just a big, giant flow of data. It looked like a huge search. So the business user was like, well, what am I supposed to do with this? He then ingested that into Splunk business flow and all of a sudden you saw a flow chart of what all that data actually said in terms of where buyers were exiting the system, calling the call center, et cetera. And then they were able to make changes through this beautiful graphical user interface. So we'll come back to that is because one would be skeptical naturally as to is it really that easy. They also announced Splunk for industrial IoT. So the thing I like about this stew and we've seen a lot of IoT announcements in the past year from IT companies. And what's happening is IT companies are coming in with a top-down message to industrial IoT and OT operations technology professionals. And we think that is not the right approach. It's going to be a bottoms-up approach driven by the operations technology professionals, these process engineers. And what Splunk is doing and the brilliance of what Splunk is doing is they're starting with the data we heard today. OEE, what's OEE? I haven't heard that term. It's called Overall Equipment Effectiveness. These aren't words that you hear from IT people. So they're speaking a language of OT people. They're starting with the data. So what we have seen thus far is frankly, a lot of box companies saying, hey, we're going to put a box at the edge or a lot of wireless companies saying, hey, we're going to connect the windmill or analytics companies saying, we're going to instrument the windmill. Well, the engineers are going to decide how it gets instrumented, when it gets instrumented, what standards are going to be used. And those are headwinds for a lot of the IT companies coming in over the top. What Splunk is doing is saying, we're going to start with the data, the data coming off the machines. And we're going to speak your language and we're going to bring you tooling that you can use to analyze that operations data with a very specific use case, which is predictive maintenance. So instead of having to do a truck roll to see if the windmill is working properly, we're going to send you data and you don't have to roll the truck until the data says there's going to be a problem. So I really like that. Your thoughts on Splunk's IoT initiative versus some of the others we've seen. Yeah, Dave, that dynamic of IT versus OT, Splunk definitely came across as very credible. The customers we've talked to, the language that they use. You talk about increasing plant performance and uptime. How can they take that machine learning and apply it to the IoT space? It all makes a lot of sense. Once again, it's not Splunk pushing their product. You're going to have more data from more different sources and therefore it makes sense to be able to leverage the platform and take that value that you've been seeing with Splunk in more spaces. So the other thing they announced was machine learning and natural language processing 4.0. They had BMW up on the stage, talking about, that was really a good IoT example. But also predicting traffic patterns. If you think about Waze, you and I, well, I especially use Waze, I know that Waze is wrong. It's telling me I'm going to get there at 4.30 and I know traffic's building up in Boston. I'm not going to get there until 10 to five. And Waze somehow doesn't know that. They, BMW had an example of predictive, using predictive analytics to predict what traffic flow is going to look like in the future. So I thought that was pretty strong. And I loved in the BMW example, they've got it, you know, it's married with Alexa. So the business person sitting at their desk and say, hey Alexa, go ask Splunk something about my data and get that result back. So pretty powerful example, really obvious to see, how do we get that, the value of data to the business user even faster? Yeah, now the problem is, I'm going to mention some of the challenges I see in some of these initiatives. The problem with NLP is NLP sucks. Okay, it's not that good today, but it's going to get better. But, you know, they use an example on stage of the Alexa, it obviously worked, they had it rehearsed, it doesn't always work that way. So we know that. They also announced Splunk, the Splunk developer cloud. They said it was three Fs, familiar, flexible and fast. Now that puts, what I love about this is, look at this is big data, actually in action. Splunk as I've been saying all week. They never use the term big data when big data was all in the hype cycle. They now use the term big data. Back when everybody was hyping big data, the big lack, the big vacuum was applications. Pivotal came out, Paul Moritz had the vision. We're going to be the big data application development platform. Pivotal's done okay there. Yeah, but it's not like taking the world by storm and it's a public company that had a decent IPO, but it's not like killing it. Splunk is now maybe a little late to the game, a little later than Pivotal or maybe even on IBM. But the key is Splunk has the data. I keep coming back to the data. The data is the linchpin of all of this. Splunk also now Splunk TV, that's nice. You're in the knock and you got now Smart TV. That's kind of cool. Yeah, but Dave on the developer cloud, this is a cloud native application. So it's fitting with that model for next generation apps and where they're going to live. Definitely makes a lot of sense. Yeah, so they talked about integrating Spark and TensorFlow, which is important obviously in that world. Stu, you in particular, John Furrier as well, spent a lot of time, Jim Kobielus, in the developer community. I mean, what's your take on what they announced? I know it was sort of high level, but you saw some demos, you heard their language. There were definitely some developers in the room. I would say, as a constituency, they sounded pretty excited. They were a relatively small number, maybe hundreds, not thousands. One of the feedback I heard from the communities is being able to work with containers and dock or something that people were looking for. They're delivering on that. We talked to one of the customers that is excited about using Kubernetes in this environment. So absolutely Splunk is reaching out to those communities, working with them. When we talked to the field executive yesterday, she talked about. Susan St. Ledger. So how Splunk is working with a lot of these open source communities. And so, yeah, good progress, good to see where Splunk's moving. Absolutely, they're listening to their customers. So land and expand, that's not Splunk's, not use that term, my term that I stole from Christian Chabot and Tableau, and certainly we saw that with ServiceNow. We're seeing a very similar playbook. Workday in many ways is trying it as well, but Workday's going from HR into financials and ERP, which is a way more entrenched business. The thing I love about Splunk is they're doing stuff that's new. Splunk was solving a problem that nobody else could solve before, whereas Workday and ServiceNow, as examples, were essentially replacing legacy systems. Workday was going after PeopleSoft, ServiceNow was going after BMC. Tableau, I guess, was going after old, tired old BI. So they were sort of disruptive in that sense. Splunk was like, we could do stuff that nobody's been able to do before. Yeah, Dave, the last thing I want to cover in this analysis segment is, talk about the data. It's the people interacting with it. We've been talking for years that there's not enough skills in data scientists. There's so many companies that are, we're going to be your platform for everything. Splunk is a platform company, but with a big ecosystem at the center of everything they do, it's the data. It's the data that's once important. They're not trying to say this is the rigid structure we talked about a lot yesterday, how Splunk's going to let you use the data where you want it, when you want it. How do you look at what Splunk does, the Splunkers out there, all the people coming to them, comparing contrast against the data scientists? Well, this is definitely one of the big challenges. I mean, to me, the role of a Splunker, they're IT operations people, they're people in the security knock, and Splunk is a tool for them to make them more productive, and they've fallen in love with it. You see the guys running around with the Fez, and that's pretty cool. They've created a whole new class of skill sets in the organization. I see the data scientists as, again, becoming a Splunker and using the tools. Splunk are giving the data scientists tools that they perhaps didn't have before, and giving them a way to collaborate, and I'll come back to that a little bit. If I go through the announcements, I see some challenges here, Stu. Splunk next for the LOB, is it really as easy as Splunk has shown, and we're going to, time will tell, we're going to have to just talk to people and see how quickly it gets adopted, and can Splunk democratize data for the line of business. On the IoT side, it's really, it's all about the operations technology professionals. How does Splunk reach those people? It's got to reach them through partnerships and the ecosystem. It's not going to do a belly-to-belly direct sales, or it's not going to be able to scale. We heard that from Susan St. Ledger yesterday. She didn't get into IoT because it hadn't been announced yet, but she hinted at that. So that's going to be a big thing. The OT standards, how is Splunk going to adopt those? The other thing is, a lot of the operations technology data is analog, right? So there's a headwind there, which is the pace at which the engineers are going to digitize, okay? Splunk really can't control that in a big way, but there's a lot of machine data, and that's where they're focusing. I think that's really smart of Splunk. The other thing generally, and I don't know the answer to this, Stu, is how does Splunk get transaction data into the system? They may very well may do it, but we heard yesterday data is messy. You know, there is no such thing as unstructured data. We've heard that before. Well, there's certainly a thing as structured data, and it's in databases, and it's in transaction systems. I've always felt like this is one of IBM's advantages, as they got the mainframe data. Bringing transaction data and analytic data together in real time is very important, whether it's to put an offer in front of the customer before you lose that customer to provide better customer service. Those transaction systems and that data are critical. I just don't know the answer to how much of that is getting into the Splunk system. And again, as I said before, is it really that easy as Spark and TensorFlow integration enough sounds like the developers will be able to handle it? NLP will evolve. We talked about that as a headwind. Those are some of the challenges I see, but I don't think they're insurmountable at all. I think Splunk is in a really good position. If not the best position to take advantage of this, why? Because digital transformation is all about data, and Splunk is data. They're all about data. They don't have to go find the data. They don't have to, I mean, they obviously have to ingest the data, but the data's there. If you're a Splunker, you have access to that data. All the data, not necessarily, but you can bring that through their API platforms, but a lot of the data that you need is already there, and that's a huge, huge advantage for Splunk. Well, Dave, this is one of the best conferences I've been at with data at the core. It's been so great to talk to the customers. We really appreciate the partnership of Splunk, Splunk Events team, grown this from seven years ago when we started 600 person show to almost 10,000 now. So for those of you that don't know, there's so much that goes on behind the scenes to make something like this go off. Really appreciate the partnership and the sponsorship that allows us to help us document this, bring it out to our communities. The analysis segments that we do, we actually bring in podcast form, go to iTunes or Spotify, your favorite podcast player, look for the Cube Insights. Of course, go to thecube.net for the video, siliconangle.com, for all of the news, wikibon.com for the research, and always feel free to reach out with us. If you've got questions or want to know what shows we're going to be in next, for my co-host, Dave Vellante, who's Dave Vellante on Twitter, I'm Stu Miniman at Stu on Twitter, and thanks so much for watching theCUBE.