SPEAKER_01 0:00

So David, there's a version of the story that you're going to tell us where you walk into a room and you accept hundreds of millions of dollars of spend and you don't rock the boat and we never meet, isn't it?

SPEAKER_02 0:18

Yeah. It would. It would.

SPEAKER_00 0:24

The qualified answer with Simon Alicia. With guest David Bruin, head of digital and online channel services at National Australia Bank in 2013.

SPEAKER_01 0:35

He chose the harder option. Give us give us some context. What was what were you dealing with? And then let's kind of bake our way out of that.

SPEAKER_02 0:44

Yeah, sure. Look, I and I think, you know, as I sit here today in 2026, I I might just go back and just remind you of where IT was at in 2011. Amazon you weren't here yet. They came a couple of years later as you were. I'm not sure it existed. GCP was still five, six years away. DevOps had only just been coined as a term the year before. So we're we're we're pretty primitive here.

SPEAKER_04 1:14

Yeah.

SPEAKER_02 1:14

We're still on-prem. We're the you know, the best we can do is VMware. Um but I personally had just come back from probably a life-changing trip around Australia with my wife and my kids who were all in primary school. I felt I needed to get a real job. You know, I was getting towards late 30s. I really love my time in architecture, but I felt I need to get into delivery. This was my first one. It was um, there was quite a big team. I'd never really done that before. But to to get to your point, I won't I won't prattle on for too long. So it felt it was presented as a bit of an intractable problem. It was a um, it was a business banking platform. Out out in the world, there's a independent industry survey, and it rated four of the four financial institutions that participated in that. And this one was a fourth of four, and by some distance.

SPEAKER_01 2:16

So it was no, it's not like four out of ten. It's like the lowest you could go was four, and this was this was meeting the requirement.

SPEAKER_02 2:24

That's right. And they only measured four, so by virtue of that, it was in there, but but the distance between three and four was the was kind of the big standout. And and and then peering inwards, looking at, you know, most enterprises do engagement scores. There was a really clearly disengaged and unhappy team structure. Product, it's really important that you're engaged, you have relationships with the product, they own the product, really unhappy. Um cost of change. That that survey by itself was a real um problem for them, the pace of change. And unfortunately, in those days when we built big Mac truck size software, in this case J2E, um it just didn't perform. So it there was it wasn't a great baseline.

SPEAKER_01 3:16

So Welcome Back from Leave was basically back from Leave. I've been selling myself around Australia, and probably not unlike many people have received in their time. It's it's something that had obviously a good origin, like it was needed by the business. It's there to serve customers, it had people assigned to it, but it wasn't going well. But the fact that it wasn't going well in the world of enterprise doesn't always yield any change, does it? Like, like, you know, we could probably point to thousands of projects that are not going well, but just keep on trucking.

SPEAKER_02 3:51

Yeah. Yeah. Yeah, I look, I think that's right. And, you know, just to just to sort of reinforce the point, is my first delivery manager role. So it wasn't a um it it was a it was a big thing for me. And yeah, I I don't think it would be unfair to categorize a lot of systems in a lot of big enterprises. You just sort of stick with it, or you you go down the replacement path, rewrite it, whatever, whatever it is.

SPEAKER_01 4:18

Try it out all together, sort of thing.

SPEAKER_02 4:20

Yeah. I didn't feel just a bit about my personality. I I couldn't sit on a problem like that without doing something about it, I guess, is is is the inert motivation for me. It just didn't feel good.

SPEAKER_01 4:30

I'd like to you weren't gonna tolerate that that scenario. That wasn't okay.

SPEAKER_02 4:36

Yeah, I I didn't feel like I was gonna be leaving work each day with a feeling like I'd done something good for people, the the institution for myself. I was not gonna learn anything. So it it was either gonna be a short-lived job, sticking with status quo, um, or we have to fix it, right? And how do we go about that?

SPEAKER_01 4:57

And so tell us about what this system was supposed to be. Like what's what does it do?

SPEAKER_02 5:02

Yeah, yeah. Look, so it really we're everyone's familiar with internet banking payments online, you know, and in that scenario, it's relatively simple. Like I owe you some cash, I get in, I know your bank details, and I pay you. Um I think in the business segment or the enterprise segment, it is that, but on steroids, like now we're running payroll, where we're doing large, complex. We might have teams to do that. In enterprise customers, we've got payroll clerks, we've got accountings, we've got CFOs, there's multi-authorization. So it's that that simple thing that we do in the retail space is extraordinarily complex to support, you know, the potential every customer that you want to have on this platform.

SPEAKER_01 5:47

And I guess it's also backed with contract law, payroll law. Like there's there's, you know, it's not like if if if you don't give me half of our lunch back in my internet but personal internet banking, you know, I give you a nasty text. This is like real ramifications if transfers aren't happening for people.

SPEAKER_02 6:04

Absolutely. I mean, just payroll, scentling payments, you know, these are real things that have impact real people's lives every day. So, you know, it's a material piece of work. And and, you know, in the certainly in the big four banks in Australia, that's that's quite sovereign, actually. So yeah, there's that's that's a high volume of specs, especially back then pre-some of the up-and-coming, you know, financial institutions that are threatening now. But back then, you know, we're really just talking the big four, right?

SPEAKER_01 6:33

So let's talk about the call that you made. Because if if I stand back as an uninformed observer, the the typical obvious choice here is well, you know, we we throw it out all together and start again, kind of a you know, start over second system effect. Or, well, it is what it is, sunk cost. Um, we've you know, we've we've come too far now. And you've gone the third way. Help us through that thought process. I'm I I really want to get in your head as to what you are wrestling with at that point. Because you've said, like, I okay, I can't go home from work feeling bad about this. So, but you've you steered away from these choices, help us decode that.

SPEAKER_02 7:11

Yeah, okay, okay. So I I think for me, the and and and to be totally transparent, I was kind of presented with these options, and you know, the person who was acting that I took over from in this delivery manager role had kind of presented with the outsourcing agreement. It was it was just ready for execution by the person who had the job. Um the the leadership at the time was had high motivation to get you know a team of really awesome developers and just rewrite the thing. Um, I think the thing that stuck out to me is you know, day one, you're walking the floors and you're meeting people, and I kind of, you know, I talked about the engagement score. I had a very real visceral feeling of people who clearly wanted to do the right thing, clearly didn't feel they had the chance. You know, so I I felt, well, I I'm just not sure the right thing is to either outsource their jobs, or or the other way is really outsourcing to a bunch of great engineers as well. Oh, I thought, well, what about these people? You know, they've they've been gunning along, they weren't part of the original builds, they've been here. They, in my first impressions, they knew what needed to be done, they were frustrated it couldn't be done. Um, they wanted to get it done. And so I really kind of sat back, went for a few walks, talked to some trusted people I knew, and tried to run through the counterfactual. So and and for me, for for I know you know me, Simon, but for people who talk, I'm a little bit metric fetish, I'm a bit model driven. So I thought, okay, well, what what what did this thing cost? What what was the total spend to date? And I won't give you the actual numbers because it'd be inappropriate, but let's say it's around about 300 million bucks. And and in large enterprises, big programs like this, that's not an uncommon. It's a big number, but not uncommon. Yeah, yeah, yeah, yeah. So um, so I thought, okay, well, what what percentage of that would be waste? And and so I just ran a little test internally, and it was a heuristic, there's not a lot of maths to this. But I I said to people who've been around for a while, the products, the leadership in tech, I said, how much do you actually think of this was a waste? And it went something like I said, 25%. No, way more than that, way more than that. Right? 50% that's 50%. Was half of it total waste? No, more than that, more than that. Um, so I said, okay, well, is it 100%? Like, was the total thing a waste? It's like a write-off sort of thing. Yeah, well, yeah, and it's an it's an interesting next step if you go down that path of of the book value, right? So it wasn't 100%, wasn't 50%. So I said, well, is 75% like is about three-quarters of it waste? Does that sit right with people? And and that that kind of sounded about right. But if you take the counterfactual of that, that means that 25% is not waste. And if you times 25% by, in this example, 300 million, that's $75 million worth of value. That is a huge, huge amount of value that I do not I did not think a group of awesome developers could just flip out. And and further to that, they would need to first get to that hurdle of $75 million before we could even start moving from there. Now remember, so $75 million worth of value to get to fourth of four before we even got to get started on.

SPEAKER_01 10:54

Yeah, yeah. So you've you've you've thrown away seven. So it's really almost a uh a 150 mil turnaround type thing. If you think about throwing away the 75 mil of value, then spending 75 to get to where you were, and like you say, and now we begin.

SPEAKER_02 11:10

Exactly. So I I know that might sound like a little bit perverse logic, I'm not sure, but that's kind of how I thought about it at the time. And and I just could not um shake that thought. And as I got to know some key people on the floor that cared um and started to engage them, I I I couldn't shake that it didn't feel right, nor was it right actually to go down either of those, you know, outsourcing paths. So so we didn't, we um I I I actually took the position to say, actually, no, we're gonna repair. Um, you know, let's just go back to the basic engineering disciplines. Where do we start? What can we do better? How do we improve? How do we measure month on month? What are we gonna do that's gonna improve ourselves to the better?

SPEAKER_01 12:02

So you've you've made the call here, and that's that's uh, I guess, the the concept here of you're a a senior leader making a decision on behalf of a team and on behalf of an institution. But ultimately the execution to some degree or to a great degree rests with them, but the accountability rested with you. And you've gone the lesser-trodden path here. What did you do, I guess, to create some form of psychological safety for yourself about that it was the right decision? And then also, how did you get others on the journey with you, be they the people on the team or the other interested parties within the organization?

SPEAKER_02 12:39

Yeah, hey, look, first clarification, I wasn't not that senior. Um it was it was deeply frightening, actually. Um I mean you raise a really good point, and I've I've never actually been asked this before, but um, I actually of that group, and there was pretty much only three or four of them. I've I've I'll give first names if if you like, but um they know who they are. So, you know, Pike, Jocelyn, David, Michael, right? And each of those people had shown the willingness to try. And I, as a people leader at the time, new to people leadership again, I said, I I'm gonna form a uh non-negotiable agreement with you. And I said, What I seek from you is to try. Now, if you try and it fails, I'm gonna wear it every day of the week. If you try and it wins, you're gonna get it. Plus some. Because and all I ask of you is to bring your best intent and just try, please. Um, so that so that was it. Uh a more complex question is my own career safety. And I and I, you know, I sort of reflected back, you know, as young, young kids, yeah, mortgages. That weighs heavily as we for most of us. And but I just had to think, well, I can't sit in the status quo. Like, I'm just gonna become totally unhappy, and and I don't think any job's worth that to me. I'm not gonna be totally unrisk managed. I'm gonna do this incrementally. Um I I'm gonna find mitigants wherever I go, but I am gonna wear that residual risk that this just could go pear-shaped on me and I might just be out of a job. And I, you know what, I'm gonna have to find another one. Um, 2011 wasn't too hard to find jobs, I guess. So I I just took a position.

SPEAKER_01 14:34

Um, but And did you did you find you had support in that position from other decision makers around the traps, or was it seen as a normally the play would be just outsource, outsource, yeah, essentially try and outsource the risk, etc., rather than try and do this slightly different approach that was a little unusual. As you mentioned, particularly in those days, it wasn't a done thing, wasn't the style at the time.

SPEAKER_02 14:56

No, no, it wasn't. And and you know, to be fair to large enterprises, I think it's hard to go against the current. And and the current in a lot of enterprises at that time was to outsource. I don't know whether you remember those times, right? So this was the default. This was against the kind of the stated direction scorecards and all that sort of thing. And so I don't begrudge the feeling that people said, well, hey, you're you're kind of on your own if you're gonna do this. Um so yeah, that that that was the the feeling. And again, I I want to be totally I don't begrudge people because that's the system they're working in, right? So that's the direction they're given. Correct, correct.

SPEAKER_01 15:39

And so it's interesting you talked about the team. And I think one thing we keep going back to is that technology is made of people, and in this uh AI excitement um psychosis, some even call it, where people are going, oh, developers won't be needed and all this sort of stuff, I think it's actually quite the reverse. And it's interesting you had you had the same team that was maintaining a system that was four out of four, um, yet it didn't stay four out of four. So what that that attention you gave to the people, because I'm sure they were coming to work trying to do the right thing in the first place, but like were you breaking out barriers? Were you changing ways of working? Like, what did you do to allow them to express that optimism and that enthusiasm in a way that actually landed?

SPEAKER_02 16:26

Yeah, yeah, yeah. Look, I I think that's where like I call signals and symbols are really in important and you know, ways of working. Again, you know, agile was pretty new back then, certainly to the large enterprises, and and we had a really convenient fulcrum between uh up until that point, waterfall, and from that point agile. So that was highly um engaging, I suppose, because that's what people how people wanted to work. So just a better, in my view, way to work. I think that's been proven now over the last 15 years. The other was just some hygiene factors, um uh desktops. Um, you know, this system took 44-0 minutes to compile back in the day. And so just loading up with some really high spec PCs, which put it from 40 minutes to four minutes, it's just you know, it's just not not working with your hands tied behind your back. And so, and then to be honest, that was the first two things and and and probably something I I think I've grown a little wiser over time, I would not may not do again, but I symbolically tore up the outsourcing contract just to show, hey, I'll I'll I'll need for you. I I uh to be fair, I was a little younger and I was a little more demonstrated.

SPEAKER_01 17:48

Look, I I guess you know, uh often we talk about the fact that unless leadership takes a position and creates that light on the hill or that vision and lives it, um you kind of got no right to ask people to follow you if you're not willing to to go out there. At least you went out there and said, hey, this is, you know, I'm burning the boats. We're doing this. Yeah. Um, but also you you backed it with action, which is, you know, and here's your extra super duper workstation that makes your life easier and better. And it's interesting how people will nickel and dime on that all the time. Like I've noticed this with, you know, can I get a bigger monitor or I need more RAM on my machine? And I think when you're looking at it from a bean counter perspective, yeah, you know, getting the latest spec'd Apple Mac or whatever, super expensive, like not cheap at all. I could give you this Lenovo for half the price with half the RAM and just get on with it. But it's it's more than just the fact that the compile time went from 40 minutes to four minutes, isn't it? There's more to it than that.

SPEAKER_02 18:44

Yeah, it was actually. It was a sign that I was willing to dip into budget to support the team. Right. And and you're right, like moving it from hygiene, a stone in the shoe, so to speak, every time you walk, to oh, hang on. Like he's not just saying it, he's matching it with some actions was important. You know, in today's speak, it'd be like capping the budget on tokens.

unknown 19:05

Yeah.

SPEAKER_02 19:05

You know, it feels like a false economy to do that. Um, to me. Now, unfettered use for non-productive purposes, no problems, but to cap the very thing seems a bit counterintuitive to me in today's speak, right?

SPEAKER_01 19:19

Well, I think it's it also speaks to the fact that there's a couple of things you've really highlighted throughout this discussion, which is firstly the Pareto principle, where you said there was 25% of value was there, like there was it was there. So that that ratio tends to just appears all now. The 80-20 just pops up in all different different ways. But also the fact that this unrelenting focus on inefficiencies and inconveniences, no matter how small they are, can yield these second-order effects that you don't, you know, because I'm I'm sure if I had to guess, I'd say that people visiting that team would notice those desktops or how fast things and they go, ooh, I want to be on that team.

SPEAKER_02 19:59

Yeah, I want to be on team. Yeah, yeah, yeah, yeah. Yeah, yeah. And yeah, look, it I I think that's part of it. And to be honest, I was learning as I went back then, but I felt and learned pretty quickly that that's what I felt my job was actually to take on that risk of mistakes. I always felt like the risk of a mistake was pretty low when there's committed people trying to do the right thing. The risk of mistake's pretty high if it's uncontrolled, wily, ill, ill-motivated people, right? But but the risk of a mistake when you've got that caliber, it's super low. So just let them have a try. Well, I think you reward them for it's the same people.

SPEAKER_01 20:39

Like often you'll talk to people and say, Oh, well, my team couldn't do that, they don't have the skills. Okay, it's like, well, how do you know? How do you really, really know that they can't do it? And yeah, I think in a lot of cases they can, they're just hamstrung. It's like, okay, I want you to, you know, do do this with one arm behind your back. It's like, well, can I have two arms and a tool of some sort, please?

SPEAKER_02 20:57

Yeah, exactly. Exactly. Exactly. And and and look, to be honest, then once we got that going, I I I want to come back to the commercial side of this, the product management. And I I do feel still today that often we it's important to index on the technology, but I don't think it's good to over-index on the technology. You can't do anything without your commercial partners, right? There's a reason the technology's there. So, you know, really building that relationship, I think, was super critical. Even, you know, to the point where the, you know, the gene of the product of the space, and she was fantastic, would come to deployment nights and you know, bring pizzas and you know, get in with us to feel the pain of having to get up in the middle of the night for a multi, you know, eight hour deployments back in the day, right? And just just that shot of empathy for the team, the you know, bringing into one team paid off so well. Um, I was always really proud of you know I I I plotted with the with my team I said you know this this person had done such a favour it's what's the gift we can give back now what are we gonna do and and what we plotted in a positive way was we were gonna do that delivery that that she never saw coming and I'd done a little bit of hunting and she'd come from a different bank and she'd been advocating to get a mobile version of this the the business banking platform for some time and so the gift we gave her is we beat that bank to it. So so so I think you know it we're all humans right how do you help each other I I think that's that that's what it comes down to.

SPEAKER_01 22:43

So um yeah and that that trust and that engagement and that that feedback that very human feedback of saying hey yeah you you busted your guts and great result like you did well and and not only that but you know everyone likes everyone else to succeed. I think what's that saying you know if it's amazing what gets done when no one cares who gets the credit. That's true. It's like if you're being if you're giving others that opportunity to elevate they'll take you with them and and you'll grow substantially like I know a lot of the folks from that core team and and the team that came from that have gone on to do outstanding things in the industry like flourished um all because of a a decision that was made not to outsource or not to redevelop pretty simple.

SPEAKER_02 23:30

Hey I I don't think it was all that to be you know you know I I wouldn't claim that for myself but you know I I do take a lot of pride in watching some of those people their careers and where they've gone and what they've done. I also take a lot of pride that it um it it got to and remains 15 years on number one of four by some distance.

SPEAKER_01 23:53

So you know so it's not only that you that you you improved it but you it sounds like the team improved it very thoughtfully because I mean it is a cutthroat industry let's make no in Australia even though it's highly regulated it's highly competitive still and the fact that it's lasted at the top for so long you know everyone's coming at the king and uh and and obviously there weren't I I don't want to use the word shortcuts but I want to say there weren't a lot of compromises made or there was a lot of long-term thinking that was going on to make sure that it wasn't like a bl like it's pretty easy to say well let's let's goose a whole bunch of stuff and get to number one for six months and then they'll be off my back sort of thing. This is a sustainable durable change.

SPEAKER_02 24:40

Yeah yeah and and I think you you touched on it before if you don't engineering enterprise we talked about the book value like what you're writing down in in that analogy $75 million. You don't get to recreate until you've hit that threshold and and it's only then to the $150 million question as you as you rightly put together but things like operations you know the support um procurement uh you know risk there's you know there's so many more factors involved in an enterprise than village I say I and I don't mean to say this dis disrespectfully but just the tech. So you know I I think it's important that people take into account the the the the context outside of the tech as well as the tech.

SPEAKER_01 25:26

Yeah you're spot on I mean there's a whole lot of conditions for success that had to be put in place be it people's default position or negotiating a position etc because there's lots of points in that project where or that approach where someone with authority could have said no or could have said stop or could have just not been helpful. Yeah no it's and and and clearly there was a an impetus across the organization to get it done and to to back it in to make it happen which I think is a good credit to the institution. Yeah absolutely absolutely yeah you talked about camoufactuals and I like that way of thinking Dave before we finish I want I want you to to give us a let's say someone's listening to this right now and they're in almost an identical position and is there a situation would you say don't do what I did so don't don't make that decision. Like is there a a frame of reference you can give us or is it a case of assess and make best effort?

SPEAKER_02 26:25

The the one thing I haven't talked about that I had in my back sleeve and this comes to the mitigation of the risk. Let's call talk about it in risk terms. So it on the surface it seemed like a big risk. When I went through the dimensions like good people you know the hygiene factors of technology the the people I was working with and to your point the culture of the enterprise which was allowed it these are all mitigants to the risk. The one thing I did have to was the right partner and I mean that in the truest sense of a partner a risk reward relationship someone in my corner who would tell me dude you're about to make the biggest mistake or yeah yeah no that's okay. So I I you know I I didn't make an unin ill informed risk and so probably the the long answer to your question is I I think every decision is a risk decision. Make it an informed one and and I I find counterfactual is a really strong you know mental model of being able to challenge what you presented as the way to go or the easy path. Someone once told me you know whenever whenever someone says hey surely back up a little bit the word surely means that it's that's probably not the right answer.

SPEAKER_01 27:38

I I I think you're right. It's right up there with how hard can it be that's right that's right yeah yeah surely we can do this on a WordPress site or you know whatever whatever whatever yeah yeah don't call me surely exactly I had to I couldn't let that go that had to be said for what um David thanks so much for coming on and being uh very open and frank about what you went through and what the team achieved um you know we're on this podcast we tell good stories and bad stories and also uh not it wasn't obvious or plain sailing but it's great to hear a great outcome and a great Aussie success story. So thanks David for coming on.

SPEAKER_02 28:18

No problems I really appreciate the opportunity. Thanks Simon.

SPEAKER_01 28:21

And thanks everyone for listening. Let your friends know the podcast is out there the qualified answer is its name. Yes with an if no with a but as I like to say there is of course a link in the show notes you can give direct feedback as well because we do love to get your feedback.

SPEAKER_00 28:35

And until next time keep on building The Qualified Answer with Simon Alicia