 This is theCUBE, I'm here with Jeff Frick, my co-host. And theCUBE is this live mobile studio. We bring to all the events. We did 27 events last year. Our goal is to do 40 this year. We invited the ServiceNow knowledge by ServiceNow. Really thrilled to be here covering this transformation of IT. We've been, Jeff and I have been talking to a number of customers. We went into the New England user group last night, hanging out there for a little bit, just talking to people. And it's just a consistent theme, Jeff, isn't it? It's quite amazing. Dan McGee is here. He's the Senior Vice President of Engineering and Cloud Operations at ServiceNow. We're going to dig into the ServiceNow Cloud a little bit. ServiceNow, unlike many cloud vendors, doesn't really lead with cloud. It really leads with business value and transformation. But behind all of that is some pretty serious infrastructure. Welcome to theCUBE. No, thanks for having me, great to be here. Yeah, so, well, let's start with the event. I mean, it's just incredible to see all these customers and their passion. And we were talking to some customers earlier today. I said, you know, you guys really, as passionate as it seems to be in the ServiceNow marketing. And they went, uh-huh. Yeah. Thank you. It's not just the Kool-Aid injection. They really love what you guys are doing for them. But as I said, behind all that, you've got pretty sophisticated cloud infrastructure. Can you talk about that a little bit? Yeah, we absolutely do. You know, and I view myself as one of our customers. You know, my entire career, I feel like I'm just like many of our customers. I've looked for a solution like this to actually make the entire product development and the operation run smoothly. And, you know, prior to my joining ServiceNow, I was doing what many of our customers were doing before they got ServiceNow. Where, you know, we sort of had these disparate islands of applications. And we were trying to tie them and sew them together with various components of technology, very, very difficult to do. We could never really easily implement the workflows that we wanted. And, you know, coming now to ServiceNow, we sort of have the best of both worlds, right? We create and we invent the product, but we actually are now using the product to run our operation and put in place exactly how we want to run things and instrument our entire operation. So what kind of advantages does that give you, say, relative to the, you know, run-of-the-mill cloud vendor? Yeah, well, you know, real-time information. You know, so many people are constantly looking for ways to put metrics in place to see, you know, in the case of us, how, with the health of our instances, how well our customers are able to use our services. And by having all of these various applications that are monitoring the entire operation, all on a common database, it makes the information instantly available and we can kind of, you know, come up with an idea today and implement it tomorrow. So what are the unique sort of architectural requirements that you feel are necessary to be successful as both a practitioner and a provider? Yeah, so we are a services provider, but I think we need to be at the highest level of availability and the highest level of reliability, much more so than a typical SaaS provider. You know, if you're just doing email and email goes down for a little while, well, maybe not a big deal. If you're, you know, maybe at like salesforce.com and on Saturday afternoon, you want to go do a maintenance and take your service down for three hours, you can get away with it because the sales guys are out playing golf on Saturday afternoon. For service now, it's a very different. Thursday afternoon too. Okay, there you go. It's a very different environment at service now. Our customers are using our product in a plethora of applications. We have them being used at the bedside of hospitals where doctors are dialing up prescriptions for patients. We need to be up, we need to be up and running at the highest level of service, 24 by seven without hesitation. So can you talk a little bit about, so as a primary customer, how do you feed back to the product groups and the service groups? How does that cycle work? Yeah, so again, we're very big on making sure those people that are responsible for the customer experience have a very tight understanding of the exact customer experience. So in our case, incident is the thing that customers use to ask us for help when they have issues. And we count the number of incidents that come in in every particular area of our product. And that information is fed back on a real-time basis directly to the people that are responsible for creating that incident. So if you're the development team that's responsible for the email infrastructure, you know on a minute-by-minute basis how things are going in terms of a customer's experience. And that tight feedback loop is then only possible with service now product. Very difficult to implement in other infrastructures where I've been. So do the sales guys then, do they drag you around saying, hey, can you come talk to my customers? They do, but I also will say that some of our best salespeople are customers themselves. So we of course reassure and we sit down and chat with customers about how we do things. But probably the best medicine for service now is to talk to another service now customer. Well, so the reason I ask is because that while that end customer may be able to describe the experience, they might not be as well-versed on okay, what's behind the curtain. So talk to me, so I'm a customer of our prospect. I love what I hear, but I'm nervous. I'm trying to sell it into my organization. There's some skepticism. Help me, let's start with the availability piece and then we'll talk about security in a minute. Help me understand how you architect availability and reliability so that I could sleep at night. Yeah, so sure, I'll start with the end result and then I'll tell you a little bit about how we do it. So last month our availability was 99.98% for all of our production instances. Very, very high level of availability. The way we do that is we actually have a completely redundant production instance for every single production instance we have. So if anything goes wrong with a customer's production instance, we can very quickly fail over to a secondary site which is in a physically different location. It's got all of their data. Everything is completely replicated and kept up to date and we can do that very, very quickly. We actually also use that same mechanism to do scheduled maintenance. So if we want to do maintenance over on site A, we can transfer the customer's active domains over to site B and then do maintenance on site A. So that's one of the biggest ways we actually maintain a very highly available environment for our customers. So when you say 99.98, what are you measuring? Is that application availability? Yes, yeah. So that's not the light on the disk drive. Right, so you count the number of seconds in a month and multiply that by 99.98% and that's how much time the system has been available across all of our production instances. And can you talk about SLA's a little bit? Is there a spectrum of SLA's? Is it sort of one size fits all? Sure, so in terms of if a customer has an outage which is our highest issue, the highest problem that we can have, those things get treated with enormous resource, right? Everybody on my staff and on Frank's staff gets an email whenever we have an issue and crisis bridges are spun up and we go after solving the problem very quickly. And then we go through P2, P3 and P4 issues. And again, we track our, the thing that we focus on most is time to relief. How long from the start of a problem a customer had, whether it be a severe issue like an outage, all the way down to just a simple question, how long does it take us to actually get the response? So in contractually, do my service levels vary? Is there a variable service level agreement or is it sort of a one size fits all? Yeah, some companies do have premium levels of support that they offer and my point of view on that is the reason they're offering premium levels of support is because their standard levels of support suck. And we don't have that philosophy. We want every customer, again, back to the starting point of the standard that our customers need us to be at. We want everybody to have the highest level of support. So every customer is a premium level support customer for us. And so do you, in your SLA, do you publish the availability levels? We do. So what is it? What do you publish? Well, as I said, so our availability is about 99.98%. So in the contract, do you guarantee me some kind of? We guarantee 99.8% in the contract. And what happens if you don't hit it? I get a refund or? Customers are due to service credits if they slip below 99.8% and if it turns out it was an issue caused by service now, sometimes our customers can do things that can cause outages because you got to remember we are, you know, our customers are developing on our platform and they change them very often and sometimes they themselves can actually have an issue that can cause an outage. Okay, so you guys will diagnose that. That's right. And so that's great. And I appreciate the candor and believe me, we spent a lot of time and we keep on looking at SLAs and digging into them. And I think in general, an SLA is a representation of how much risk the vendor is willing to take. And it sounds like, I like you right on. When you see an SLA that has, okay, you can spend $180,000 a year extra for the premium platinum service. That's a real red flag. I'm impressed that you guys don't go there. Talk to me about, oh, one of the questions I have related to sort of availability, where's the data? Can you talk about your global presence? Yeah, so we have, we're up to 14 data centers worldwide right now. And the location of our data centers is largely driven by the needs of our customers. The Swiss banks like to have their data resident in Switzerland. The Canadians like to have their data resident in Canada. And so, there you go, right? So we have data centers across all the major continents. We're building out Brazil as we speak. And we're going to be moving into Asia early next year as well. So our data centers are pretty much all over the world. And then we have a redundant pair in each major location. Okay, and that's all sort of, you guys own all that infrastructure. We're in Colos, right? So we don't actually own the physical real estate, but we own the equipment that's in those data centers. We administer it. It's our employees that manage it and take care of it. All right, now let's shift gears a little bit. Talk about the other, you know, the other evil twin of the cloud, which is available, I mean, security. So, again, I'm a prospect, but my folks in my organization say, no way, we can't outsource to the cloud. It's too dangerous, security is a problem. Help me, you know, sell internally my peers on why your security is comparable or even better than mine. Yeah, so we do a number of things. There's obviously a number of standards that we comply to already. And we just recently got, I think we're one of the few SaaS providers in the world that now has the federal government authority to operate for FSMA level two standards. And we also are constantly doing our own testing, our own penetration testing with third party vendors trying to attack us to try to expose some of our issues. I think security is one of these topics where you're never done, you're never complacent. We are constantly working on this, constantly trying to improve it. But what we'll do with the customer, you know, many customers come from different industry segments where they have very different kinds of requirements. Pharmaceuticals have different requirements than banks. Banks have different requirements than the Department of Defense. And so we will work with each of these industry segments and go through what we're doing, have a chat with them about what they need. And most often there's a heavy degree of overlap between what we're doing and what they need. Sometimes we need to do a little bit extra to make an individual customer feel better. A lot of times what you hear from customers that complain about the cloud as well, they may feel as though or agree that your security is better than what they can get. But sometimes they're concerned about compliance and the ability for them to meet the edicts of their organization and prove that to their board. So talk to me about, let's say things like audits. Can I come onsite? I mean, for instance, Amazon's not going to let me come onsite and they have good reason for that, you know, physical security. But that causes problems on the other end. What's your philosophy in terms of, for instance, onsite audits? Let's start there. We do allow audits. We have customers that do them from time to time. Not everybody does them, but we realize some industries, you know, they're required to by their regulators. Again, the pharmaceuticals is a great example of that. And, you know, we do allow these customers to come onsite and do audits. One of the things we're trying to do with many of these customers is again, pharmaceutical example, we're trying to work with a consortium of pharmaceuticals so that they can have common needs and perhaps drive one audit on us as opposed to every customer driving a different audit. Make it a little bit more efficient, a little bit more successful. Yeah, how's that working? They're actually meeting at this event and working through that and making good progress. That's good. Now, how about something as mundane as how I define an incident and how we report incidents? I mean, it's, as you say, every customer, not every customer, but many customers have a diverse set of needs. So will you work with me to determine what is an incident, how it gets reported, how frequently it gets reported so that I can comply with, as they say, my corporate edicts? Yeah, we certainly will. You know, and, you know, we don't have too many issues in that area, actually. I think, you know, again, we're part of sort of ITIL and the ITIL standards, and within those standards, you know, the definition of a P1, P2, P3, and P4 is pretty clear. We don't get much pushback on those kinds of things with customers. You know, I think what customers are more interested in is are you working with Vigor to resolve our issues? Yeah, sure. It's more, you know, what's really behind the infrastructure as opposed to some sort of, you know, contractual thing that's written down. It's more the people and more how we're behaving that's selling our customers, not necessarily what's written in a contract. So if you use an ITIL level of granularity, again, a lot of cloud providers don't. It's just sort of, this is what it is. This is how we report it and take it or leave it. So, okay, you passed the test. Yeah. Great. That's very good. That's a question. How has the reception to a cloud service changed in the large enterprises over your time? Yeah, so, you know, I think, like a lot of things, it's started out with skepticism and concern and worry and the way customers sort of address that concern is they would start with something simple, something that they deemed maybe a little bit lower risk for them. And as they build confidence in us, as they build confidence in SaaS, they're putting more and more of their infrastructure into the SaaS environment. And, you know, if I've been with the company now for a couple of years and I remember the early discussions were a lot of these kinds of things you're asking me about today, it's much less that way today. Customers have a lot more confidence and I think that confidence is only going to continue in time and, you know, SaaS is going to become more the norm than the exception. So where have the customers, or the conversations now shifted their focus to? Are those still the big, you know, kind of projections or are there some new things that? You know, we still run into some customers that would like to talk about how we do SaaS and the behind the scenes kind of thing, but more often than not, things are moving more towards where you see Fred talking, right? What can you do to help me? Tell me how I can automate this process. Tell me how I can take advantage of ServiceNotTechnology to make my life easier and make our company more efficient. And that's where we want the conversations to be. You know, we want sort of the backend operation to be relatively invisible. It's always on, it's always there a little bit like the power company, right? And we want to spend most of our time helping our customers really solve their business problems and make them more efficient. And that's what we see happening. That's great. There's an IT practitioner and I'm going to wrap up. What, if anything, do you learn from the big web guys, the hyperscale people, the Amazons, the Googles, the Facebooks? Are they instructive to you or are they just sort of, you know, too heavy engineering and PhD-oriented or are you able to sort of borrow from some of their best practices? Actually, they are very instructive. And you know, many of our employees actually come from those companies. And so we, you know, we learn every day, right? We learn from the employees we hire from many of these companies. We learn a lot from our customers. You know, we like to think that we're pretty good at what we do, but we're not so good that we don't have something to learn every day. So we're always open-minded. We're always talking to customers about how they run their operations, comparing them to how we do things. And it's a very productive feedback loop for us. That's like, I think it was Bill Joy. I'll probably hack the quote, but he said something to the effect of, no matter how smart the people are at your company, this somebody's smarter working in another company. That's a great time. All right, Dan. Well, a great story. Really appreciate the candor and the really humble nature with which you guys approach this. Always willing to make improvements, but passing the, as I say, the test of scrutiny for relative to many cloud providers. So well done. Congratulations on getting it done. And of course you're not done. So good luck in the future. We'll see you next year. All right, thanks very much. All right, everybody, keep it right there. We are going to be cutting over during our break to some content from Google I.O. We've got a team, a CUBE team from SiliconANGLE on the ground there. So we'll be running some content there. And we'll be back here after our break at around 11 o'clock Pacific time with Craig McDonough, who's the director of product marketing for ServiceNow. And then we go deep into our customer experiences. And we're meeting with Maritz, CareWorks, FICO, Yale University we'll be on today. And then later on we get Beth White coming on and the CFO of ServiceNow, Mike Scarpe-Haley. So keep it right there. We'll be back after this break and after this word from Google I.O. This is the CUBE. We looked at all the programs out there and identified a gap in tech news coverage. There are plenty of tech shows that provide new gadgets and talk about the latest in gaming. But those shows are just the tip of the iceberg and we're here for the deep dive. There's a difference between technology consumers and those who live the business day today. And our viewers recognize that. The market begged for a program to fill that void. We're not just touting off headlines. Our goal is to provide you with a story but we also wanna analyze the big picture and ask the questions that no one else is asking. Our guests aren't just here to provide commentary. We work with analysts who know the industry from the inside out. The tech business isn't new but many networks treat it as if it is and really barely scratch the surface on technology coverage. We follow the expansion of the cloud and the evolution of big data. We're covering new enterprise from startup to IPO and every move in between. So what do you think was the source of this misinformation and so you mentioned briefly there are several other. If that's the case the player. I like to think of us as a companion to theCUBE. We're here every morning trying to extract the signal from the noise. Where theCUBE excels in event coverage. We're working to bring that experience to you consistently every morning. We use the top stories of the day to provide you with breaking analysis so that you.