 Live from Las Vegas, it's theCUBE, covering Knowledge 16, brought to you by ServiceNow. Here your host, Dave Vellante and Jeff Frick. Welcome back to Knowledge 16, everybody. This is theCUBE. theCUBE is SiliconANGLE's flagship program. We go out to the events. We extract the signal from the noise. Frank DeSantis is here, he's the CTO of the Hartford and Sarum Dube is here, he's the director at KPMG. Gentlemen, welcome to theCUBE. Glad to be here. Knowledge 16, what's happening for you guys? A lot, we're getting a lot done here, learning a lot. This is our, my first year here, actually. And thanks for the promotion, by the way. I'm the director of service management. How are they at CTO? Oh great, sorry about that. Come on theCUBE, good things happen. You'll be the CIO before you know it. I know. So this is your first year, Frank. Yeah, we had put ServiceNow in a year ago, actually during last year's conference, so. Well how's it going? It's going really well. It's been quite a year. We've got a lot done and got a lot more planned. So we'll talk more about that. But Sarum, what's your role at KPMG and? So I'm a director with RCI Advisory Practice. We are the practice that kind of owns our ServiceNow Alliance and I'm a solution architect. I focus more around the build out of the solutions and generally clients like Hartford and especially when we're talking about large enterprise scale and implementations, there's a certain way to go about them and I work with the team that kind of develops our methodology for it. So you guys were, Frank called you out as one of the key partners or day one. Yeah, we use KPMG as the implementation partner. So how was that decision made? Actually take us back, Frank, to the decision. So you had some guys here last year kicking the tires. Is this real, talking to people? Yeah, we had a strategic goal to SaaS, right? And so we were looking on the market for, first we were looking for the tool providers and we narrowed it down to the two biggest SaaS providers for service management tools. We did a bake off on that and then we chose ServiceNow because the decision was obvious from a functionality and a experience perspective. And then we looked at the major partners with ServiceNow and we narrowed it down to three and at the end of the day we picked KPMG. And you were replacing an existing tool set? Yeah, we were replacing the BMC remedy tool set, yes. So when you bring in ServiceNow, are you retiring systems or are you sort of the people hanging on to the legacy? Yeah, it was a different experience in the sense that they talked about customer experience. It was very low with the product we were placing. So it was easy to bring in ServiceNow because all of the users are very frustrated with the old technology. And so we did a lot of demos and got people excited about the new tool. And so that's how we made the decision and implemented it. And from the SaaS perspective, do you have kind of a broader SaaS directive with your whole suite of applications and this just happened to be one piece of it? Yeah, we're insurance company, right? And we don't want to own infrastructure and we don't want to own applications. We just want to lease space, lease seats. So ServiceNow is an ideal solution. Prior to that, I implemented a SaaS, our PPM, our project and portfolio management system, which is not ServiceNow at the moment, it's an HP system. But this whole strategic direction is to try and SaaS as much as possible. Because we think of the stag, we talked about it a little bit, right? Hartford is a conservative, you know, kind of old school insurance company. So for you guys to really take kind of a cloud, a cloud-dative approach and start to swap out on-prem stuff with cloud shows pretty progressive move. It's progressive, we've been extremely cautious, you know, a lot of security concerns around data and everything else. So it actually took us, you know, we tried to go to ServiceNow like three years before the time we went to it, but they were not a publicly traded company. They were a very small company. And from that conservative approach, we decided not to do business until last year. Obviously, you guys know they've grown dramatically. They're publicly traded, so we were a lot more comfortable with it. And Sarabh, what role does KPMG play in these types of implementations? So I think we started off pretty much at the top of this one. We started with helping Frank and team build the business case for, you know, switching over from BMC to ServiceNow. And I was thinking when Frankie were talking about the whole consumer experience piece being so better, it was two days with workshops that we did just to see if people would get interested and try to get their requirements. The response we got was overwhelming. We had two days, eight hours back to back, packed up with people coming in with their requirements of this is how we want to do it, but we're not able to at this point. So it started from the top, helping build the business case from a qualitative and quantitative point of view. I think even the internal rate of return was huge on the trend. So big that you think it was embarrassing, right? It just didn't work. One of those deals, right? Don't show this to anybody. They'll not believe it kind of thing. Yeah, so. It kind of a whiteboard exercise in terms of if you start from scratch, here's how would you like your system to work or what you've shown for two days to get them so excited? It was not so much of a start from scratch. It was more around, let's start through what you have and what you want to be at and let's find out how you'd get there. So the you have part was the two-day workshops that I was talking about, but where do you want to get and how do you get there? Those exercises then happened over a period of time and you kind of work together to define the full roadmap of how this would get implemented over time as well. So we had two separate engagements. One was to help us do the business case and the other one was actually implementation because we had, in our case, we believe it's a sizable implementation that with 30,000 users and... 30,000 users. With 30,000 users, we had to convert. We did it in a nine-month period with KHBMs, whose health we had like an army of people onshore, offshore and did a lot of people change management on that. And it was extremely successful implementation. It's... So before we come back, because I want to peel the onion on that a little bit, but the business case, can you talk about, give us a high level as to what that looked like and when you get these situations where IRRs are telephone numbers, a lot of times conservative companies will just strike out anything that's soft dollars and go right to the hard dollars. Can you talk about that a little bit, add some color to the business case? Well, yeah, kind of. We're like a share numbers. No, yeah, you don't have to share numbers, but just again at a high level. It was extremely compelling and we didn't use any soft numbers whatsoever. But part of the business case was displacing a couple of systems, but it was also reversing an offshoring support decision. So we actually developed an in-house support team for the system, because that turned out to be far less expensive to support it onshore than offshore. So if we looked at the whole package, we're talking about around a 2 million run rate a year savings and... And the big chunk of that came from you guys in order to service customers or whatever you were staffing up offshore, because it's cheap. And then all of a sudden, wow, we could automate. We could automate and onshore... Sometimes offshore looks better than it does because you've got a lower rate, but you're stuck to that 40 hour work week versus doing it onshore. And we find that the people working on this system just love working on it. So they're far more productive than working on the previous system. So it's a quality aspect, the morale aspect. Now... Absolutely. Whatever apples to apples kind of a thing. Sorry for interrupting. No, please. One of the things that we focused more on was if you're going to compare... If you're trying to build a business case, it's more than just the numbers that appear obviously in front of you, right? It's not just the license cost, not just the cost of people. But when you have to go apples to apples, you have to look at if I was to develop everything and getting in this solution that I'm going to need. What would that have cost me otherwise? The apples to apples part in this case, I think that played a huge part as well. So really look down the road, not necessarily today, but what I'm going to need down the road. Yeah, we were kind of stagnant because we were very mature as far as having individual process owners for every process. We had multiple IT organizations. We brought them together over the past five or 10 years. We brought our process teams together. So we were very mature in our process structure, which gave us the ability to move faster. But we were stuck with the old technology because we could only go so far. So even though we knew what our ideal process was, we couldn't implement it in the old technology. And that was one of the dramatic improvements moving the service now. KPMG showed us how, yeah, you can do that in the new system. So you had to fit the process to the tool as opposed to the tool being flexible enough to fit whatever process you wanted. Did you have to do a lot of mods to bring this in? No, we really didn't do a lot of customizations. We have a large service catalog. So there's a lot of development. I wouldn't call them customizations around that service catalog. And we have a fair number, about 37 integrations to systems because we didn't implement this as an isolated system. We embedded it at Connecticut to our ERP systems and other systems. So it's part, it's core of our whole IT systems. Now I got to go back to something that Sara was saying, the apples to apples comment. That assumes that somebody was going to sign the check for that those, you know, that cause, was that the assumption going in? Did you have the support of the CFO or whomever? You know, okay, we're going to do this no matter what. Now pick the best path. Is that right? I think it was a five year plan that kind of, so Hartford does a five year business case. A rolling five year architectural plan. So when we built that, it showed that within maybe the next two years, the numbers wouldn't match. But the moment you cross the fourth year onwards on a rolling basis, the apples to apples starts playing a very big role in that. So yeah, the expectation was that if you didn't do this, three years from now, you're going to start losing money on it. I love business cases like this that can justify themselves on hard dollars. You can strike out the soft dollars. It's always easy to get approved. But the reality is is when you go implement, I wonder if you could confirm this or deny it, but the soft dollar value oftentimes dwarfs the couple of million bucks that you're going to save. Is that what you found? Yeah, no, absolutely. We found considerably more savings than we expected from a soft perspective. But it's very hard to tie that to dollars. And in our particular company, we can't even talk about, well, I avoided 10, 7, 1 incidents this year, which would have cost you X number at all. Throw them out of the meet. Get out. So unless you can give them a code that it's coming out of, you can't talk about it. Sign up for more budget or less budget in that case. So, okay, so now, so it's been a year, you said? Yes, yeah. I got to ask the Mulligan question. And if you had a Mulligan, what would you take it on? Or would you do anything differently? No, actually not. We decided to do the Big Bang, do everything at once, because it amazes me that some people say, well, I'm going to do the CMDB first and this other thing second. All that stuff is so integrated, you really can't do it separately because you draw out your problem and you actually cost more money to implement it. So we, in nine months, like converted everything, incident problem, change, asset management, release. And the big discovery. Internal comms effort and training effort and we're hearing that over and over that you really need to have a support behind that. We spent a lot of time and money using KPMG's change management resources and we can't say enough about that. They did an incredible job, people change management because we literally had to train at different levels, but all 30,000 employees. And you had a flop over date? On a particular day, you switched? So people in change, they focus on behavioral change management and training. So they were involved from day one. It's not like we had to bring them in halfway through the project. They were involved from day one because they had to understand what were they checking for, right? Organization as big as this, there's definitely some detractors. So the whole readiness assessment, understanding where this could potentially become an issue, catching that ahead of time and solving all those things, all of that was part of what they were doing. So substantial project, 30,000 people. When you do a conversion like this, I guess it's not like a cobalt to mainframe to something. But do you have to freeze anything, freeze code or freeze process? Well, we freeze three months into this project when we start, we freeze the current environment. So we're only processing incidents, no enhancements, no changes to the current environment. In one aspect, it's kind of easier because we're banding in an old system. So it's not like I'm having to upgrade it. So I'm building a new one and then moving people onto the new one. So we were able to keep the old one up and running for three months to run off tickets. So you didn't have to convert a ticket. So that was helpful. And then the only thing you lost is the new enhancements. Right. Not really, because the new enhancements that were being requested were now going to the new tool. So as... The new tool wasn't up yet, right? Yeah, but the requirements would... I think the timeline was anyways, right about, if somebody gave a requirement, it would have taken three to five months to actually build that requirement anyways. Right. Okay, so it was a net neutral. So it didn't affect the business case negatively at all, that freeze. No, we had a task from our CTO. Yeah, you can do this if it pays for itself. But finish that sentence if you would. If it pays yourself in five years, he's going to say no. But so what kind of... In a five-year CBA, it's we're making money at the first year. After the first year, we're saving money. So it paid for itself within two years. So you're breaking even that fast. And the net present value is enormous, right? With that's something of that size. Interesting that you chose... We treat by your comments about, we just did it a whole house. Because it's somewhat antithetical to this notion of you're here today agile, take it in small chunks, et cetera, et cetera. But sometimes that just doesn't make sense. It makes sense. If you're going to do it, do it. Well, that's the whole thing with service management. As you can see, it's totally integrated. It's hard to take a piece out. Like if I say I'm going to do change management first, talking to some people, well, how can you do that without your CMDB? Right? You can't. Well, we can do it manually, but you don't get the savings and you put more risk into the situation. If you don't take at least those core service management functions in a bundle and move them in together. And we're not talking about completely leaving aside the agile methodology. Once the first big bang happens, then the approach becomes more of a hybrid agile because at that point, you start to request your incremental updates, the new enhancements that are coming in, start following in that cycle. The first big bang, because the scenario that we had had to be done in one strong integrated way. Again, you're leaving a solution which was built over what, seven, eight years. So going from there, you have to bring in a big bang approach. But then from that point onwards, you have to show the agility and say, we're going to do this very quickly. You're going to have shorter releases for enhancements as opposed to the first one. We're in a very agile. We do a release every week. It's like the presentation the other day. We have a train that leaves every Thursday night. If you're on it, you're going into production the next day. You're pushing code every week? Every week. Yeah, so you're right. That's not antithetical to agile. You know what it is? We talk a lot of startups and they don't have the money to do a big bang. They don't have 30,000 to take a piece at a time. So that's true. So what about that whole scrum and agile? That's a sort of cultural change that has been affecting your life at the heart. It is, and we're looking at that. And we did, like Sarb said, a hybrid approach to implementing it because it was such a large project. But from a service now evolution perspective, we're totally in an agile environment and how we release code and how we have our testers. From a Hartford's perspective, we're looking at agile and DevOps in a larger sense right now, outside of service management, but obviously it will integrate with service management. So we see the developer conference going on and how does, and I think I'm correct the Hartford has pretty big development team and pretty high quality. And so how do they fit into the whole service now? Because these guys talk a lot about low code or no code, but if you're intense code, you can go for it. What are the developers saying about service now? Well, the developers use service now to process incidents, changes, release, to do application support from a development perspective, they're outside of service now as part of our DevOps initiative. We're gonna be integrating with whatever PPM tools and whatever development tools we align on because right now that's another initiative that's going on outside of service management, but we're part of that because we need to support it. So we're looking to rationalize our development tools and testing tools and getting more DevOps perspective from our application, customer application development teams. But all of those teams use service now on a daily basis to move code into production, to process incidents. So when I think about the application, what the application portfolio must look like at the Hartford Dan McGee this morning out if you saw his keynote, but he threw out an IDC number that said sometimes organizations have thousands or tens of thousands of apps. You're one of those organizations, clearly. We give you a amount of, somewhere between 1400 and 2000 apps. Yeah, okay. And then you can sort of think about those in suites and you got claims and you got your agent systems and et cetera, et cetera. We've heard a lot yesterday in particular, even some today, about service management touching the third of state. Is that resonate with you? We're integrated. So we use a different tool. The architects use a tool to manage that application portfolio as far as the attributes. This is what the application does. These are the processes. These are the owners. We've integrated that into service now. We have a nightly feed into service now where we take those applications and we create different environments and we relate them to the infrastructure. And then we connect it on the back end and it goes directly into finance for monthly charge backs. So we've integrated the front end and the back end, but by doing that within service now, we also can track incidents against applications, incident against service and not just infrastructure. You said charge backs. You're doing charge backs, not show backs? We're doing charge backs. We did show backs prior to service now and now we're doing charge backs. And we do it based on the relationships in the service now, CMDB, between the applications and the infrastructure components that comprise that. We gotta go. I got like a million more questions, but all right, both of you guys give you last words. So just your thoughts on Knowledge 16. Frank, your first time here. Well, it's very impressive, right? It's amazing size of a crowd. I don't think I'll ever eat food here again because the walk is too far. Sorry, I'd rather pay for my lunch. But I like some of the keynote thoughts and service now really needs to get out there and talk more about being outside of that core service management function and more about the end to end integration. I think that's a huge opportunity. And Sara, what's the end of the bumper sticker on the show from KPMG's perspective? This is my fifth knowledge. I've been here pretty much as long as you have been here. Fourth for us. Fourth for us. So I was at New Orleans the first one. But I think we're seeing that evolution of clients going from, let's start with ITSM, then let's turn it over. It's almost like the ownership of the platform, ownership of service now inside organizations moves from the ITSM group to the platform group. And that's when the third estate really kind of starts playing out and they start thinking about what else can we do with this? And what Frank said about this is we can do much more than just ITSM ITSM, it's just a starting point of it. We're starting to see that turn happen with a lot of our clients as well. Great. Well, Sara and Frank, thanks very much for coming to theCUBE and sharing your story. So it's a good one. Thank you for talking to us. All right, keep it right there, but we'll be back with our next guest right after this is theCUBE. We're live from Knowledge 16. See you right there. Three once in a while. A true break.