 Live from Las Vegas, Nevada, it's theCUBE. Covering Knowledge 15, brought to you by ServiceNow. Okay, welcome back everyone. We are live. This is theCUBE, our flagship program. We go out to the events, and I just have to signal from the noise. I'm John Furrier, my co-host Dave Vellante. We are live in Las Vegas for ServiceNow's Knowledge 15. No 15 is the hashtag. If you want to join the conversation, go to crowdchat.net slash no 15. We'll answer your questions. It's all live thread. Go jump in and chat. We'll take questions. Our next guest, Dan McGee, COO of ServiceNow, just gave the keynote, Chief Operating Officer, handling all the build out, all the backend, all the big iron in the cloud, creating the awesomeness for your platform and your customers. Welcome to theCUBE. Good morning, how are you? So how do you feel on the keynote? I mean, third day party, started late, was it packed, was it people strolling in late? What was happening? I was great, you know, it's been a busy week. There's still a few days to go, but I was actually pleased to see how many people were at the keynote this year. It was wonderful. One of the things we were commenting on our intro this morning is looking back at the two days of broadcasting live here and looking at the history of our immersion in with you guys, covering you guys, we talked to everybody. And the theme that's coming out is enterprise grade, ServiceNow success. I'll see you guys are very successful. And cloud native, I call it the DevOps mindset. You're seeing the confluence of DevOps in the enterprise where cloud is the key. One of the clouds that people say for the startups. But now that the cloud is mainstream in the enterprise, it has to be a little bit different, but all the benefits of DevOps without all the hassles of building DevOps. I want to get your take on that trend. And as you're operating the company, what do you believe in that premise? What do you see? What are some of the things in ServiceNow that's unique that you could share with the audience around the cloud piece? Yeah, so I think in fact, this is just in case you asked me a question that doesn't go so well. Just going to be well armed. Yeah. Where's my water? You don't get any. Hold on, let me tweet that. I have a phone call to me. Hold on. So that question was okay. You're safe. Oh boy. Let's get ready to the jokes. Okay, so the whole point is that people want to put more stuff into the cloud. The enterprise wants to take advantage of the public cloud, but they have these questions and these concerns about the reliability, the security, the safety, transparency of the cloud. And what we're really trying to do is meet that need and communicate to folks that we get it, right? We really hear what customers are asking for. They are raising the bar for the kind of infrastructure they want to deploy on, and that's what we're trying to deliver on. Well, you put forth the premise. Most public clouds are ill-suited for the broader enterprise adoption. That's right. And you make it pretty bold statements. Last year you showed a glimpse of some of the availability numbers. So why the big push now is because you're getting so many questions from customers or you just have some muscles to flex. Yeah, well, I think customers are actually going through this natural transition. They're doing the same things that you and I were concerned about as we started using things like Gmail and Yahoo and stuff like that, where we were initially kind of concerned and worried about trusting the public cloud with our stuff and then we got more comfortable with it. And this is in fact what we do here from customers today. And I think too many companies have just sort of taken their cloud infrastructure for granted or they've sort of done what people 20 years ago did when they established a cloud and what we're trying to point out is that's fundamentally the wrong way to go when you're trying to address the enterprise where people need to have some sort of control over their instances. They don't want to have this thing just be completely opaque and invisible to them. So the next generation of the cloud, one of the things that in the opening keynote with Fred's keynote was this notion of, look where we've been, the founding of ServiceNow, what we've delivered to customers, pushed back on the memo from Sequoia, we punched back to the market and put some great traction. And it was the setup for the next 10 years, right? So you can see that the tea leaves like, okay, next 10 years of ServiceNow, you're in charge of building that out for the company. What is some of the insights that you can share on how you're investing in this new generation of the cloud? And what does that mean for the customers? I mean, as the customers look at ServiceNow, you've got a lot of happy customers, they're all here, the party last night was really a testament of how cool ServiceNow is and relevant you guys are. Next 10 years, how do you look at that as an executive leading the charge? A lot of pressure, but this cloud generation, what is unique about this and what would you share to customers on what you're doing? I think of a couple things. I think the first thing I think of is where we started the conversation, which is the enterprise needs to have a rock solid foundation that they're going to trust, that they can actually put their data and their trust in this cloud is actually going to be there for them. But then above that, I think what you're seeing from us is trying to really enable a partner community that can go out and develop applications on this platform. I think ServiceNow is going to continue to invent and innovate our own applications, but as you look forward to the future, more and more of the applications are going to be developed by the community at large. So we have CreatorCon this year as an example of some of the stuff we're doing there, some of the platform improvements that you saw talked about during Fred's keynote or a big piece of that. But those are really the things that are top of mind for me is how can we enable creative people to develop great stuff and do it on an infrastructure that they don't have to be concerned about or worried about? So a couple other questions we got during your keynote. So you showed that sort of taxonomy of the consumer cloud, the departmental cloud and the enterprise cloud and you had ServiceNow and you compared it with a number of other big clouds, in particular AWS and Azure you had at the top, which is kind of an interesting comparison. Somebody said, well, that's kind of interesting, comparing one single app with some infrastructure as a service. So why did you choose that? Is it because there's such prominent clouds or maybe you could talk about that a little bit in clarify? Yeah, well, I think it starts with mindset and I think when those folks envisioned their services, they envisioned a service that would be delivered to the entire enterprise, right? They didn't start with sort of just a departmental focus. They didn't start with just a consumer focus. They built something that was intended to actually be delivered to the entire enterprise. Now, they live at a different area of the stack as we talked about Amazon is provisioning bare metal and you need to go in and provision your own OS and things like that, Azure a little bit higher up the value chain, but fundamentally, we're equivalent in our mindset and our focus. I think it's easier to develop an application on ServiceNow and get it to market a lot more quickly but the point I was really making with that slide is what's the mindset, what are they really trying to solve for for the enterprise? Obviously at this event, you're making a lot of the multi-tenant versus multi-instance. You laid out sort of why multi-tenant is less suited for the enterprise, things like failover. You mentioned that you can't do a failover with multi-tenant. I'm not even sure you can, you can't even test failover. You don't failover is really what your point was and you guys do that. What's the trade off to multi-instance, Dan? And maybe you could help us understand that. My experience is always sort of a way to skin a cat. What's the trade off that you guys face? Well, first off, just to clarify, on multi-tenant you definitely can do failover. It's just such a risky event that people choose not to do it and so you're always afraid to sort of push that button. Presumably the same thing with testing. You can't really test it because you can failover fail back. It's too risky. That's exactly right. I was speaking with some folks that are in that lower tier that are at the show, the consumer focused folks. And again, those folks are trying to get every sort of last ounce of performance per dollar out of their infrastructure. And so in those cases, they are still using some very heavy iron sort of solutions that are fundamentally multi-tenant because in the consumer world, you don't need to treat customers individually. And I think that works well for those folks that are potentially in the consumer space or delivering components of their consumer solution. But again, for the enterprise where you want to have control, transparency, some individuality and a little better fall isolation, it's really not the way to go. Multi-insist is much better. Okay, so you're saying there is no trade-off there that there's not a, is there a cost trade-off? I mean, it's directly. You know, I think when you're dealing with the scale of the folks like the Facebooks of the world, the eBay's of the world, I think those are really smart folks and they've chosen to go with multi-tenant and they could work for them. There's really no downside for them. I don't question their architectural choices. I think they're pretty smart folks, but they're not trying to deliver the same kinds of services that the enterprise really needs. So when did you make that decision to go multi-instance and was there sort of, I'd be very interested in the reasoning behind that at the time, take us back. It couldn't have been a 100% lock, we're going to do this and everything's going to work out. It actually was and it was because some of the early architects at ServiceNow came from those very early cloud companies and they lived through those experiences of the multi-tenant architecture and they learned sort of the hard way and I think we were fortunate enough to have them and smart enough to listen to them. So it was dogma on their part in a sense. It absolutely was. Good dogma. Yeah, to do it this way, not that way. And in fact, you still see this today where companies try to go and deploy clouds and they think it's a simple thing, they think it's an easy thing and they go and do it and then they have a difficulty with it and they have to go get some help. Healthcare.gov is a notable recent example of this kind of thing. So we were fortunate enough to actually benefit from these folks. That's a great point about the innovation comes from these web-scaled companies like that had to build out their own stuff from scratch, clean sheet of paper. I also want to kind of talk, you mentioned the failover, I want to talk about security. We was just on a chat this morning with RSAs going on in San Francisco. The big conversation around obviously transparency, sharing information. You guys are very transparent companies. One of the messages we're hearing throughout the company and your customers. I love that about you guys. Talk about the transparency aspect of you guys and the security evolution of how you guys work in that, I want to say gray area, but I mean, you have cloud, this security concerns. Talk about security in the cloud for service now and then how that plays into some of the things around transparency and threats, detections, et cetera. Yeah, so security is probably our primary focus. Even if it's up all the time, and if it's reliable, if it isn't secure, it's not going to be comfortable for customers. So we just spend a tremendous amount of time trying to make ourselves secure. But we do it in partnership with our customers, right? We don't claim to be the only experts in the world around security. So we involve third parties to help us penetrate, test our environment. We even involve customers in doing the same kinds of things. We subject ourselves to audits of various kinds of standards committees. You know, our goal on the security side is just to be the best we can possibly be. And you know, we've got a lot of our own internal folks that are pretty billion at this, but we're not afraid to take help from, and advice from anybody on that. But you're proactive on that front. I mean, that's what you're saying. Yes, absolutely. And just to clarify, there's a lot of confusion in the cloud. I can't, for instance, go and audit Amazon's, you know, data centers, although they will now allow me to use their auditors, I guess, to check things out. So there's movement there. What's your, what's service now's position on that? Can I bring auditors in? Absolutely can, right? You know, we ask that you schedule it, right? Then it's not surprise visits, you know, and we ask if you're going to do a penetration test, you tell us about it, because if you don't, we'll probably detect it as an attack and we'll shut it down. But you know, we do it in concert, but we're not, you know, we're not shy. We love a third set of eyes looking at us all. And you showed, it was a good slide you showed with the, you showed the threat matrix of a typical data center and then you sort of simplified it with service now. Maybe you could talk about that a little bit. Yeah, you know, so my story on that is, you know, I think we're pretty sharp, but I don't claim to be any more brilliant than anybody else on the security side of the story. But I do claim that the problem we're trying to solve is much simpler than what the average enterprise solved. The average enterprise is a very complicated infrastructure. They're dealing with email, they're dealing with SAP or financial information, HR information, applications specific to their line of business. They have a very complicated environment. Lots of back doors, lots of front doors on ramps, off ramps. ServiceNow has a much simpler environment. We only do one thing, and that's host service now instances. So securing our environment, detecting threats, it's a much simpler problem. And as a result, I think we're going to be a more secure solution for the average enterprise. One of the things I want to chat with you about is to develop our traction. Obviously 1,200 people signed up for the CreatorCon. And it kind of reminds me of that slide that Fred put up about the first ServiceNow event, get not enough chairs, and then all of a sudden people show up at the door, kind of gate crash, and get trampled, come in, get seats. So talk about the dynamic there, because this relates to some of the things you're investing in. And develop our community is not an easy thing. One, two, there's operational issues around having this new private instances, which you guys has been a big hit here from the folks we've been talking to. So talk about that and how that relates into what you're doing and investing in in terms of infrastructure cloud. So one of the fundamental visions that Fred had when he created the company was to make an application accessible to sort of the common man that could actually do programming. And I kind of count myself as to one of those folks. So look at somebody like me, I've got 30 plus years of experience in industry. I have definite opinions about how certain processes should be run. They're not right or wrong compared to somebody else, but they're mine. They're the things that I've grown up with in work. In the past, if I was going to deploy a particular application, I'd have to go hire a pile of consultants. They'd come in with a trailer, right? A long standard, a statement of work, a big description of what we were going to do. And then I still wouldn't get what I would want, right, I get this combination of almost what I wanted, but not what I wanted, wouldn't work. Then I'd have a support nightmare. I'd have this thing that now I need to have these consultants around forever to continue to support this thing. ServiceNow enables guys like me that are maybe really good with things like Excel macros or JavaScript and these kinds of things. I can actually go in and make my vision happen in a much simpler way, time to value. Taking that solution that I've had in my head for 30 years and actually delivering it to my company is something you can do with ServiceNow. And that's the energy of CreatorCon. And that's why I believe so many developers are so excited about ServiceNow. It's easy to get something up and running. So what you're saying is the foundational service as your consultant example means, you have to essentially build the factory of data and infrastructure to make a decision, that platform, that foundation over and over again. Here you're saying, okay, foundation's set. You put your IP and ideas on top of it. That's right. And then you build on it and leverage, kind of like an open source ethos, build on the shoulders of others. The platform comes with lists already built in, forms already built in, workflow already built in. All that machinery is stuff that in another environment you'd have to write yourself. That all exists and you just need to sort of connect it together and make it look like you want. And yes, we and a lot of our partners out there provide really applications as a great way to start. Well, you're 30 years in the industry. I want to get your perspective on things. We've been having a debate in Silicon A. We'll keep on around the startup community. Evaluations are high. You're seeing a lot of startups getting a billion dollars. Here comes water. Okay, it goes in water. Hold on, I think a phone call. And then we're going to speak great. So what's the difference between spending money and investing? I mean, because a lot of the, some of the critics and the investors can be saying, these companies are small. They have to build out from scratch. They're just spending money. They're not investing. It's a little bit of a nuance, but like, how do you, how would you explain that to an investor? They're just saying, identifying, coming that's spending money and someone that's spending money to invest. Because some are saying that it's sort of hard to go from zero to, a trajectory of infrastructure overnight. You guys are in the middle of that and you're investing. Yeah, I don't know much about what other investors are saying and stuff like that. I just know what it's like living and working at service now. And I can tell you we are investing to keep ahead of our customer demand. If you look at the infrastructure that we're building, you look at the data centers that we're placing in regions where nobody else has data centers. We're doing that because customers are there. Customers want us to be there and we're going to get more and more customers over time. So we're investing to sort of sustain the growth of the company because we think we have just gotten started here and now's not the time to sort of cut back and try to turn profitability ridiculously high just for no reason. We are building the best infrastructure and the best product for our customers today. So follow up on architecture. We're talking earlier about the consumer guys, everybody calls them the hyperscalers and they've made choices. We have seen your TAM continue to grow as you've envisioned new markets, new spaces. Do you feel like you've got the architecture that can support that kind of TAM as you grow and what gives you confidence in that? Yeah, I think we do. As I say in my keynote, I don't know that we're going to be chasing video game space because I don't think architecture is for that. I don't know that we're going to be chasing storing very large data files like a YouTube service or something like that. Those aren't the spaces we're going to be at. But when you're talking about managing work across the enterprise and helping sort of all of the internal departments across the company to get more effective, more repeatable, we absolutely have the architecture that will scale to that. It's a time to value thing you mentioned and also agile mindset, having to solve customer problems fast. Yeah, but I'm inferring, you didn't make a trade-off between availability and scale. Is that right? No, absolutely not. Yeah, not at all. And so it brings me to my availability. John's going to, his eyes are going to glass over now, but I loved that chart that you showed last year. You showed even more data this year. You're the only company that I know that actually measures availability at the end user level the way it should be, not the light on the disk drive. So you've got to be proud of that. That's great. I know there were some folks that weren't in there. I noticed no Oracle. Now it's early days for their cloud. They're known, and I'm just picking on an example, but they're known for having pretty high availability. If you're, and a lot of your customers are Oracle customers, would you be concerned as an Oracle customer of putting your data in their cloud? You know, I don't know. I don't know because they don't publish their data like we do, right? I don't really know what their real performance is. I just know that a lot of those companies and those examples that I showed that claim very, very high availability, yet when I go log on to their site, I can't get in for a little while. That's just sort of the experience that I think people are getting kind of tired of. And I would be really, really happy if all the cloud vendors actually reported availability as we do, because then I think everybody would really know what's going on. Well, I think you're setting the standard, and I would encourage everybody to really push their vendors for that standard. But so now let me ask you the flip side of that question. And I tease Amazon a lot about this, more than tease. I poke at them a lot about the SLA. What's the SLA with ServiceNow? How do you stand behind that availability figure? What happens if I don't achieve that availability? Yeah, well, first thing, whenever there's an outage, we get all over it. There are contractual terms where if it dips below a certain amount, you get a service credit. So everybody asks for that. That's pretty standard in the industry. We have a number in our contracts too, but the reality is our performance is so much above that it's where that happens. It does happen from time to time. But we're all about just making it better and better and better. The data that you showed this morning is essentially the SLA that I can expect as a customer. Yeah, with using that methodology that you... That's absolutely right. So talk about the growth plan for you guys. You're in charge of building out the infrastructure footprint, et cetera. We see a lot of successful clouds like Amazon, for instance. Really the global equation's huge, right? So what's the global plan for you guys? And obviously there's data issues around countries, Ireland, Germany, and then you got Asia Pacific. What's the plan for ServiceNow in terms of that global perspective? Obviously cryptography is a big part of it. How do you look at that investment plan? What's the mindset for you guys? Because you guys have a plan. Now you have a lot of customers that are going global. What's the plan there? Yeah, so again, it all starts with we want to make customers comfortable with putting their data in our cloud. So what is that driving? Well, that's driving what they're asking for. That drives the 16 data centers in the eight regions around the world, right? We're putting data where they want to be. And that's also driving some of these encryption features that customers have asked us for. So many customers have sensitive data. Not all their data is sensitive, but sometimes there might be a social security number or something else that might be stored in our database. They want to be able to encrypt that. And some of those customers want to know that we have no ability to unencrypt that. And so that's what the edge encryption feature that we announced this week is all about. So we're going to continue to come up with features, place data centers in places where it makes sense, all with the idea that we want to remove the worry about putting your enterprise data in our public cloud. And share some perspective. What's it like to work at ServiceNow? Take us through day in the life of what's going on in the company. What are the days like? What are the people like? What's going on in the meetings? What are you guys arguing about? What are you guys fighting for in the direction of the company? Obviously growth is a big part of it. You've got growth. And I hope you don't kind of, holding on to the rocket ship of ServiceNow. Next 10 years is in motion. What's it like there? What's the culture? And how would you describe it to folks out there who may or may not know what's going on in ServiceNow? It's a great company, but it's the mindset of the folks at ServiceNow is what's the next problem around the corner? What's the next thing to solve? We don't take a lot of time to really celebrate yesterday. We're always worried about tomorrow. And so that's what sort of drives this incessant concern and worry and maybe even a little bit of paranoia. We wake up every day, figuring out a way to make life better for our investors, better for our customers and more reliable. I was going to ask Frank this and Fred, but I forgot to, but I was going to ask you, every company's got that unique cultural DNA feature, Moore's Laws Intel. What's the ServiceNow key thing? It's transparency. We tell it like it is. We tell our customers how things are going. If we screwed up, we say we screwed up. If they screwed up, we helped them figure out what went wrong and we're that way with our employees as well. So it's transparency and honesty about what's really going on. It's the only way for us to get better. I got another under the cover question. I'm sorry, John. No, that's fine. So John and I were talking yesterday about the database that's under the covers. I said it's MySQL, which in that I go on Core and I see confirmed it's MySQL and I think we've probably talked about it before, but I seem to recall that there was a no SQL add on or aspect of what you guys were doing whether it was Cassandra or I couldn't remember MongoDB. What's under the cover? And then in Quora, it says I can have an option of Oracle if I want, which was news to me. Help us clarify sort of the database strategy. Yeah, so in the early days of ServiceNow and the very early days of ServiceNow, the customer actually got to choose because back in those days when folks were just sort of getting comfortable with going into the cloud, right? They had all their own database admins that were expert on infrastructure. And when they didn't have a lot of confidence in the public cloud, they kind of dictated a lot of what the public cloud would do. So in those days, they would say we want you on Oracle or we want you on this or on that. And over time, people have gotten a lot of confidence in ServiceNow and they don't really dictate that anymore. And in fact, we don't necessarily even really allow it anymore. But we still do have a handful of sort of Oracle customers that started off with us in the early days and we're quite good at supporting Oracle and we're going to continue to support those customers on Oracle. To grandfather those customers. Absolutely, but going forward, we're definitely building on MySQL. But fundamentally, the actual database that we sit on isn't really the thing that makes the difference in the service. We could probably make any one of these things work. We just sort of landed on MySQL and we want to continue to grow with that. Is there a NoSQL component? We do have sort of some interesting options that we'll roll in to a particular customer's instance if it will benefit them. We have things like read replicas and other kinds of technology that we invented. But again, we do all this behind the scenes. That's how we achieve some of the performance numbers I showed you earlier. So customers just don't have to worry about it. If their transaction rates are getting pretty hot or they're doing things that might benefit a slight tune to their infrastructure, we're going to do that behind the scenes and they don't even need to know. But you must get this question from customers. They're IT guys, they want to know, right? Sometimes, you know, I think the conversation used to be a lot more about that. And honestly, now the conversation is shifting a bit more to the security side than some of those questions you're asking. Dan, my final question is getting close on time here is you guys have been very successful. I talked with Fred and Frank about it, is that you're disrupting and innovating at the same time, which is very rare. And so a lot of people want to know the secrets of success. And we hear from your customers all through theCUBE this week, is that it's a bold step. It's like, will the parachute open it? Because they're forging new ground, the new use cases, new capabilities are showing innovation opportunities and workflow, new use cases, new workflows that could be automated, all the greatness that you guys have provided. So you've been very successful in taking a bold step into a new world, basically a new generation. So question for you, what is the learnings that were magnified over the success of ServiceNow? In this, as you guys have been growing super fast, a lot of customers happy, you're kind of going into these new areas where you're innovating and it's working, right? So what learnings would you share with the audience that have been magnified in that process? Yeah, I think in my piece of the puzzle, it's just been one step at a time, right? We have climbed some, what others would have imagined were insurmountable mountains, but we've gotten there. And we've done it by just taking one step at a time, knowing where we're going and getting our head down and making it a little bit of progress every day. And you see us continuing to grow and continue to be successful. And there are new challenges out there, right? Some of the security things we talked about. One step at a time, we introduced new features one year at a time, one release at a time. And I think you're just going to see that continued methodical approach to delivering what folks really want to see. Is the manager philosophy just smaller checkpoints, a little bit short in the mile post, if you will, on the execution? Interate quickly and learn. You mentioned the learning from stakes kind of thing? Well, certainly, in my part of the organization where we're doing things, yeah, we don't bring our hands for too long about, is this going to be a really successful product or not successful product? We're much more biased to, hey, let's get something that works and get it out there and see what people think and then let's build on that and let's react to that. So it's safe to say you're data driven. Absolutely. And feedback driven. We listen to what people have to say about what we're doing. All right, Dan McGee, the Chief Operating Officer at Service. Now here inside theCUBE, he's got a couple of waters there. I'm going to end it right now. I should get my other suit. Thanks for watching. We'll be right back after this short break. This theCUBE live in Las Vegas, be right back.