>> Man: Live from Boston, Massachusetts, it's theCUBE, covering Red Hat Summit 2019. Brought to you by Red Hat. >> And we are back live in Boston as we continue our coverage here on theCUBE of Red Hat Summit 2019. It is our sixth year here at the show and this year obviously some huge announcements. A significant moment it's been for Red Hat, we heard from Jim Whitehurst a little bit ago. Stu Miniman, John Walls, we're now joined as well by Niall Fitzgerald, who is the GM of IT Application Architecture and Design at Spark NZ. Niall, good afternoon, or I guess good morning still we're in an Eastern time zone. >> Yeah it's the middle of the night in New Zealand I'd say. >> Yeah, so Spark NZ New Zealand. Tell us a little bit first off about Spark NZ. What the folks back home are doing right now, work-wise, and your role with the company. >> Yeah, so Spark is the largest provider of telecommunication services in New Zealand. All the traditional type of services you'd expect, mobile, broadband, et cetera. We came out of the traditional kind of post office, so we've a lot of heritage, and about four years ago we rebranded from Telecom New Zealand into Spark. To represent that we were changing from being a telco into much broader range of digital services. Our purpose is to help all New Zealanders win big in the digital world. >> Niall, step back for a second. Talk to our audience that might not know the telecom industry as well as you, I've been an observer and participator in the industry but you know back in the dot com boom it was like limitless bandwidth and we're gonna do all these wonderful things, and cloud and digitization, have put some new opportunities as well as stresses and strains on your industry so, you know what's going on and you said you rebranded? >> Yeah, look, I think it's well-known it's been a tough last few years for most telcos in the world. I was listening to Red Hat talking yesterday about 60 consecutive quarters or more of growth, I don't think there's any telco in the world that probably has the same story. Like most, we're facing kinda decline in all the traditional revenues like voice and text and things like that, so we're all having to kinda rebrand ourselves and deliver much higher levels of customer service. People expect the same levels of service from us that they do from Amazon, Google, and everyone else. In Spark what that means to us is we've moved into lots of new things as you said, things like ICT, we're now very big in cloud, we've recently launched a Spark Sports brand and we've got streaming right to the key events like Formula 1. We're going to stream the Rugby World Cup, which is a massive event for New Zealanders, so looking forward to seeing that and Ireland on the all blacks in the final in September this year. So yeah, lot going on. Tough times but forcing us to keep changing every year. >> And so, about these changes that you're making whether technologically based, let's just deal with that. What is that ultimately going to do for you in terms of better customer service delivery? So, you've got inherent challenges, you've talked about them at all, that the world's changing, how we use this medium, this communication opportunity is changing, and you've been just a little behind the wave, hard to keep up with it, so rapidly changing. How much of a challenge is that? And then how are you going to address this going forward? How do you stay relevant? >> Yeah I think we're lucky in one regard because if I look back about five, seven years ago we were like most traditional telcos. We had a spaghetti for want of a better description of systems, and then we had all was multiples of everything, at the time we had 19 integration layers and 10 billing systems and it wasn't uncommon. But way back in 2012 we actually embarked on a massive transformation program, and we spent five years consolidating all of that infrastructure so going into about 2017 we were very lucky in that we had a massive foundation laid already, so what that then enabled us to do was to actually push away calls from our contact centers into mobile apps, into digital adoption. We've been a big embracer of things like big data and robotic process automation as well to try and take cost out of our industry. So, I think we're quite well placed. Now that allows us to do things like innovate new products for our customers so we bundle things like Spotify and Netflix. It allows to introduce things like Spark Sports brand, which we couldn't have done five years ago before the transformation We just wouldn't have been able to enable these things with our existing kind of legacy IT estate. >> So how's open-source play into all this for you? >> Yeah open-source, I suppose our first foray into open-source was when we went to start embracing big data and automation. So we started using things like Hadoop and various other things and our entire platform is based around open-source. We changed to an IMS network recently and we started embracing things like OpenStack, and then it really took us to a new level recently when we started working on Red Hat's Fuse, and OpenShift we started implementing that. >> Okay, so the OpenStack show for many years, the last few years we saw the telcos coming in specifically for network function virtualization or NFV. Is that what you're using in that space? >> Yeah, we are. Interestingly, at this conference I've heard a lot of people talk about OpenShift and OpenStack, obviously, particularly in the telco game. We actually came out a bit differently from the application space. So we had an integration platform that we had put in through this transformation phase which had served us well, and was connecting all of our 40, 50 systems together. But it was coming up to a life cycle event, and we decided we'd look externally and see had we options beyond just upgrading it. So we started looking around, and we effectively found Fuse, and in bringing in Fuse we then brought OpenShift in, which is quite different to what I've seen from a number of other people, they're bringing in things like OpenShift and building on top of it. We did it the other way around, you know? And we did it primarily for cost reasons, you know? >> Yeah, so talk a little bit about that impact of Fuse and OpenShift, what that means. Were you already down the containerization journey, or did that help drive >> Niall: No, no some of that modernization? >> That's exactly what happened. If I'm honest we hadn't really explored containerization too much because we had come to the end of our kind of transformation journey. Open-source and containerization wasn't around when we went through that. So we kinda needed some really core reasons to move on, so, yeah effectively what happened was we looked at Fuse, I was gonna say primarily for cost, but we were looking for something that we could migrate to where it makes sense. We were looking for something that wasn't a massive lift for the people who worked in our integration already, so they could be rescaled into it, and interestingly we turned agile recently which has changed the way we look at the needs of our systems. So our old integration platform, if we needed to deploy a change we had to take an outage, which was fine when we had a centralized IT department who deployed once a month and took a two hour outage, but when you have 20 tribes all developing features in isolation and they wanna go straight through to production, if everybody took an outage then our systems wouldn't really be up very often. So one of the key things that we were looking at for our new integration platform was can we deploy hot and can we scale? So that's basically where Fuse came into us. >> Okay, so can you? >> We can and we do. Still a little bit nervous about pressing the button mid-day and doing stuff >> Right, simultaneously and thinking this has really gotta work, right? >> Yeah then normally, >> We saw it today though on the demo stage, on the keynote. You know, simultaneous operations going on. >> No, we do it, and they normally don't tell me when they're doing it they just do it and tell me it worked afterwards, but no it's actually been really successful and you can imagine connection 40 or 50 systems together is effectively the equivalent of about 2,000 API's and we managed to migrate, we're about 70% of the way through. But we've managed to migrate those without actually impacting the systems that use them and that's probably been one of our most successful IT projects that I've seen. >> It's funny, you said we were towards the end of our transformation journey, and of course I think we all understand, it is just, I might've reached a marker in my journey, but it needs to be a continuous process. And you went through an agile transformation. So bring us in a little bit. Organizationally, what happened there. Some of the good, the bad, and the ugly of agile, 'cause I mean agile's always an ongoing thing. >> It is, yeah. So about the start of last year we started to think about agile and the need to change our ways of working. And we looked at a number of models overseas, and companies like Spotify and various banks, and we settled on a model of chapter and tribes. And we took about six months in looking at what that meant for us as an organization and all of the things that we needed to change. Everything from, people's contracts to people's titles. We got rid of all complex titles and moved down to simple things like Developer, Tester, et cetera. We had to train our people in agile so we ran boot camps for over 2,000 people. We had one with 500 people attend. We had to review all of our processes and see where we had centralized things like IT governance or procurement. How do you actually manage this when you have up to 20 different people effectively, or tribes doing their own developments, so over a period of about six months we went through all of these. We started with a concept of some forerunner tribes so we could figure out how this thing actually works, you know? And get some lessons. And then on the first of July last year, about a 2,000 people in various buildings packed up their stuff in their desks and moved into a new world, into their tribes with different working spaces and different collaboration areas and all the tools that we need. So, yeah we're about nine months down that journey now and it's been good. >> How many total employees? >> We have about 5,000 in total. >> 5,000, so you had 500 at one time. 10% of your workforce in training at one time. >> That's right, yeah. Absolutely. >> How do you keep the wheels on the bus rolling? Because I mean you're asking people not only to learn new skills, but learn them in a new environment, and learn them literally in a new place. I mean that's just massive change and I think, we're human beings. We're creatures of habit to a certain extent. You had to hit a lot of bumps along the way. >> Yeah, so one of the key things we did upfront was we said the operate part of our business, which is effectively things like our contact center, our sales staff, our service desks, we will not go agile with those on the first day, because they operate in a slightly different way of working. The people in our stores, et cetera. So we had a concept of agile light and agile heavy. So we kinda parked them for a minute so that we wouldn't do exactly what you say and let the wheels fall off the trolley. And we took to people that were the IT developers, the product development staff, and all of that, which came to just over about 2,000 people, and we firstly flipped those 2,000 people and put those through bootcamp. But even as you say, scheduling the boot camps, we made sure that we always had the right people on the ground and we would schedule smaller boot camps for them later if we needed to do it, but yeah. >> So nine months in now. You talking to your peers, if they're gonna go through. Any key learnings, what were some of the most challenging things that you ran into? >> I think probably the major one is that agile at its heart is a way of working, and despite the name it's actually quite prescriptive in how you should work, you know? When you pick up the agile book it tells you all the ceremonies you need to run and the processes that you need to run as well. And I think you need to be pragmatic in how you implement it because there are so many different flavors of agile. The one flavor, even with an organization of Spark size, it doesn't work. So the tribes and squads that are building out new products compared to the tribes that are doing things like upgrading systems, they will work in different ways. So I think the first thing is be pragmatic, take the goodness and the intent of agile, but implement it in how it works for you. And there's some other practical considerations, like prior to being agile we had quite a large number of our technology partners were based offshore in India, and you know it's quite difficult to run a 10 AM stand up in New Zealand setting the priorities for the day and the sprint plans, when, you know, four members of your team are asleep in India. You know, they're missing out on all of the goodness and the collocation and the sharing, so one of the things we had anticipated that, so luckily enough we had moved a lot of those people onshore in advance of agile, you know? But it is a big cultural change for everyone in the organization, not least the leadership teams as well. >> John: Well you got through it. >> We got through it, but there's no going back. >> Absolutely, no you're in the deep end now. Well, Niall, thanks for being with us, we appreciate the time joining us here on theCUBE, and I think that an Irishman is always welcomed in Boston. >> Thank you very much! We've been enjoying the hospitality. >> Yeah the door's always open. >> Thank you very much. >> Thank you very much. Niall Fitzgerald, joing us from Spark NZ. Back with more here on theCUBE, you're watching this live at the Red Hat Summit 2019.