 Rwy'n meddwl, rwy'n meddwl i'r micrwyffrindig. Ynno, ddweud i i ffraeg yma. Mae'r amser yn ystod. Rwy'n meddwl i'r ystod. Ego yw fynd i'r ymddiolol a'r ystyried ymddiol yn ei wneud? Rwy'n meddwl i'r Bank o New Zealand. Mae'r Bank yn New Zealand. Mae gwybod y gwybod yn yr adrodd y tîn. Mae'r cymdeithas ar gyfer yn ymddangol. I'm the one on the left, by the way, in case you're thinking, the one on the right is my little boy. So, thanks a lot for having me at Agil India. I've had a little tour around Bangalore on Monday and found some fellow Kiwis, which was great in the market. So, that's always good to see. Now, this is my Agil ego, right? It's pretty big, right? My Agil is pretty big and I'm going to talk more about my ego and how that plays into the playbook a little bit later. So, I want to talk about a Bank of New Zealand, how we're set up. I started working with them around about three years ago. And being said digital, it's about 300, 350 people, something like that. Part of the bank, the bank as a whole is 5,000 people. And being said digital itself is set up along the Spotify snapshot model. You all roughly know what the Spotify snapshot is. Tribes with groups of squads or self-organising teams around them. And a tribe, fundamentally, is centred around a product, a customer, some jobs to be done. And we're structured like this. So, we've got a partner's tribe and they look at business digital banking, so online banking, mobile banking. We've got what we call consumer wealth. They look at retail banking and they're split up into sub-tries, what we call Hapu. Hapu is a Mādi New Zealand term for a sub-tribe. Then we've got what we call a platform's tribe and they look at engineering, some of the stuff that the other self-organising teams consume. And then we've got our creative services. It's a bit more like an agency model, but it's shared across all these other teams. And this includes UX designers, UX researchers and practice coaches of which I'm one. And then we've got performance and capability and they look at data analytics and things like that. So, each one of our teams has got a product owner. And when I first came in, there was a bit of an awkward relationship between product management, UX and Agile. And it was a bit like a three-way handshake. It was a bit of an embarrassing handshake from New Zealand World Cup a few years ago. And it was a very difficult relationship. And when I first came in about three years ago, one of the senior designers says, can you come and speak about Agile to some of the rest of the design team? And I said, sure, yeah, I'll do that. So I went along and I started to talk about Agile and it all sat like this. And I kind of start to read a bit about body language and I could tell they were a little bit switched off about Agile. So I said to them, hey, what do you think about Agile? And they went, I won't paraphrase exactly what they said, but it was along the lines of Agile is a bit rubbish, right? It's an engineering thing. It's what engineering teams do. It doesn't apply to us from a UX perspective. We hate it. And I went, ah, OK, so what do you do? And they said, we do the lean process. Wow, that sounds really good. And I said, can you tell me what the lean process is? And they said, well, we focus on a customer and we look at that customer and we think about what outcomes that customer wants to achieve. And even through a few steps, we look to deliver a minimum viable product to that customer. And I thought, well, that sounds a lot like Agile, right? Or what Agile should be. And there was some sort of disconnect going on here. So I teamed up with one of the senior designers and collaborated with her. And we started to introduce Lean UX. And Lean UX from Jeff Cothell fundamentally, you know, it's a mash-up of Lean UX, Agile, all these sort of good stuff. What Agile should really be? And one of the things that we wanted to focus on is, really, and this sums it up from this great quote here from Melissa Perry, product management is not about building things, right? It's not about building something efficiently. It's about solving problems. How could we make these teams? How could we make people within engineering care about the customer, care about customer problems? So we introduced Lean UX as an experiment across a couple of teams. And that was great. And, you know, what we wanted to do was make sure that the whole team has ownership of this. We focus on creating an outcome. And we do the smallest amount of work possible when we start measuring the change, right? Sounds like really good Agile to me, right? But we didn't kind of use that word. And we started to rock on with this. And we started to go through and think about, you know, again, this comes from Melissa Perry. Well, we started to think about, right, number one thing we want to focus on when we start when we introduced Lean UX is we're going to focus on problem identification, right? What problem or opportunity do we want to solve? And once we've established what we think is a problem, we want to actually make sure that it is a problem, right? We want to validate it's a problem. We want to reduce the uncertainty. Make sure that we're being effective. And then we start validating what solutions we've got. And then we start optimising on that solution. And very early on, well, we started running this dual track approach. You've heard of dual track, dual track development. Kind of a coin term, specifically coins that's been run with Marty Cagan and Jeff Patton. And really it's about two teams, the two teams running in tandem, right? So we've got discovery happening in tandem with delivery. And we've got, you know, UX. We've got product owners. We've got discovery team working closely and hand in hand with the engineering team who's getting this, you know, working efficiently to get this thing out. So we've got discovery, which is a focus on fast learning, validating that it's the right problem, validating that the customer actually wants this thing and we're being effective. And then we've got our delivery teams focusing on predictability, that's a difficult word to say, and quality, making sure we get the right, efficient things out to the customer in a quality way. And we've got some success on this, right? And one of our best successes was when, you know, we had a project manager for the tribe come along and he was like, well, this linear work stuff, how do we know it's going to be successful, right? What are some of the outcomes we need? And he was a bit cagey about it and he thought, well, aren't we just going to spend ages doing naval gazing and making sure that we're actually getting something out there into production. And very early on we got quite a good success. We were basically migrating old PC banking onto new internet banking, right? And we had a few features left. And one of these last features was a very specialised audit logging tool. And the team sized it up and they said, well, we reckon it's roughly going to be 52 weeks' worth of work. 52 weeks, right? And you know what traditionally what we probably would have done is we would have cracked on and built it, right? And we would have built it in 42 weeks maybe and we would have turned around and we would have gone, yay, we've done it, right? We've been agile, we've delivered it efficiently in 42 weeks. We are awesome. And we might have gone for a soft drink or something and a bit of cake and a celebration. But what we did is we started with, you know, some assumptions. And our first assumption was customer wants it. We were going to laugh about that right customer wants it. But what do you think we did? What do you think we did first? We did some discovery, right? We went and talked to some customers and we said, do you want this thing? No, no one wanted it, right? And we looked at the analytics and no one was using it. And so we went back to the state card as we were not building it. And I went, why? And they said, no one uses it, no one wants it. And they went, great. Off you go then. Go and work on something else. And that team spent 52 weeks going on working on something effectively and something that the customers actually valued. So we had a really big success, right? And this success, Snowball, we had a really good success story, which meant that they wanted to roll a linear ax out across all the other teams, right? So from two teams up to about 30. Which is great, right? We thought awesome success. We've done it. Two teams to three tribes. And very quickly success became confusion, right? Confusion because, yeah, we were kind of, you know, the two of us were helping and being this product conscience for these two teams. And now we spread across 30 teams. There was confusion. What is dual track delivery? What is this linear ax stuff? Can you come in? And the two of us, we just couldn't cope with these 30 teams. So one of the things that we found was dual track was becoming dueling tracks, right? All these different egos, one of them was mine, right? There was like the lean thinkers and, you know, there was the agile thinkers and there was a design thinking and UX and UX were turning around and saying, you're stepping on my toes and you're doing this stuff. And it was getting really confusing, right? And we were getting really worried about this. So that's it. So I hope that people would see me on the left, right? I hope people would see me as this wise old sage, you know, telling people about agile and about lean UX and what we should be doing about the customer. One of the reasons I grew a beard is so it could look wise, right? With a bit of white in there. But you know what? People saw me as on the right. You know, I came across as Mr Shouty agile pants, telling them how to do it, what to do it, telling them how to do their jobs and that wasn't my intention, right? And this was causing me a bit of consternation. So I love this quote. You can't read the label. You can't read the label of the jar you're in, right? And I was in this little jar. I can see my head up the top there. I was in this jar and I couldn't really think what was going on, right? I was getting confused. I was in the weeds and I thought, well, I need to step back and think about what I can do to start spreading this knowledge across other teams and start working with people. So how could I become the conscience of the team? Because for me, as a practice coach, as an agile coach, as anyone, I am the conscience of the team, right? I don't want to start questioning their answers. I want to question their thought process and be their conscience. And I thought about this quote as well, right? This chap, David Husman, unfortunately passed away last year. But another quote from him, he talks about, we are often making a mistake of conflating, scaling and spreading, right? We didn't need to scale this stuff. All we need to do is spread the knowledge across other people so we could replicate the good stuff that was happening in these individual teams. One of the things that we found when we first introduced Lean UX was the power of language, right? The power of language really united us. Often what we found is we would end up with focusing on the language of what we do, right? The frameworks like Scrum or this or Camman or whatever, Engineering or UX. And what we found was as soon as we started to focus on the language of why we do things, rather than what we do, it brought us together. So this is some of the language that we found really important, right? Who is the customer? What's the problem or opportunity we're looking to solve because we'd have people come up and saying, we don't always have problems. Sometimes we do have opportunities, fair enough. What assumptions do we have, right? Assumptions is a really powerful word because it can feel the weight coming off a team. It's not a requirement. You've got to do this. It's an assumption. Hey, here's an assumption. What experiment are you going to run to validate or invalidate that thing? What are you going to do to make sure that it's the right thing to do? The other thing we found here was about success, right? Success outcomes. And we talked a lot about failure in our job, right? Hey, we failed, right? And that actually makes me a bit cross on, I get cross. It's my white beard again, right? But I get cross because for me, we've succeeded, right? We've succeeded in invalidating one of our assumptions. That's not a failure. That's a success. Celebrate it. So what we started to do was celebrate these invalidation of these assumptions that we held before, right? We didn't go off in 52 weeks and waste 52 weeks of money. We were going in effective direction and building stuff efficiently as well. So what we wanted here is to make sure we understood why we did the stuff, right? How we do stuff, why we do stuff. So we made sure teams understood why rather than just the what. The other thing that worked really successful when we first brought in Lean UX is that collaboration, right? And as a coach, I collaborated with senior designers, senior UX designers, right? And that really helped. And there's a pattern from Linda Rising and she calls it a change pattern, a collaboration pattern, right? And if you want to introduce some change, what you do is you go and you might say, hey, I've got this great new idea, right? For Lean UX. You like your Lean UX, right? Would you come and help me trial this thing out and see if we can make it work, right? Automatically, you're part of that change, right? Yes. We're part of that change together and we can start that snowballing. So what we wanted to do is to make sure with this success we would collaborate with other specialities across the group, right? Other specialities and they could bring their expertise, right? It wasn't about me going to them and telling them how to do stuff. It was about me making sure that we could collaborate. They could bring their expertise and we could mash it together. So we found across his idea of creating a playbook, right? And I got really... Again, I was getting cross, right? I got really cross about seeing these playbooks which are about 1,000 pages long. Nothing playful about that. It's just cut and pasted stuff, frameworks for the internet. And we wanted to create a playbook that was our playbook that was a high level over you about how we did things, right? How we delivered products, how we did product development. So our playbook test kind of thing in the air was it had to be understood by anyone, right? So anybody who picks it up can understand it. Well, that means lots of pictures, nice and easy to understand. No more than 30 pages. Just sounded about right, 30 pages, you know? Didn't have too much heft. Self-explanatory and it was endorsed by all the skill sets and they contributed to it, right? It wasn't just one person, it was everybody's. And it had to be enjoyable to read. So these are the five lessons that I learnt from creating this product playbook. And number one, if it doesn't fit in your head, don't use it, right? So often we use terms, we use these words, we use these frameworks and we don't understand what they mean, right? So why use it? And I'm as bad as everybody else on that, right? And I had someone come up to me the other day and they said and yeah, we've got a problem in the tribe and I think what we need is I'm going to introduce a new operating model. What's an operating model? I don't know, right? And, you know, often we just take these things as a road and he said, well, expectations. Expectations between the teams and sort of the tribe leadership. OK, why don't we just call it that? Why don't we just call it some expectations mapping between the two? Oh yeah, OK, so we did that, right? And one of the things that I found very early on with this stuff in the head was people found it really hard when the simple two lines to explain what dual track development was, right? Dual track development felt more like dueling tracks, right? It felt more like, oh, we are the discovery line, OK? And we will go away and we will discover and we will hand over discovery to the development team and the development team will go away and they will build and off they go. And it felt very separate, right? It felt more like dueling, fighting between the two. It didn't feel as it should that these two were coming together. And I was getting really confused about this. So what I did, everyone should have a coach, right? I've got a few coaches and I went to one of my coaches and I said, I'm really struggling with this, right? These dueling tracks. This is not quite working in my head. I'm trying to map this stuff out and I'm trying to explain this in a way, but I can't. And he sighed and he looked at me, my coach and he said, I think we need to go and get something to eat. I thought, well, OK, we'll go and get something to eat. There's got to be something here, right? So I went with my coach and we went to get something to eat and we sat down and he took me to his favourite restaurant I paid. And there he is. There's my coach in the top right and my son Owen. And he started playing with this lazy Susan. Do you know what a lazy Susan is? Yeah, like a little, it's like a thing you spin round. You put food all in there in the middle off to get them in Chinese restaurants. You get food in the middle and you spin it round. And he's playing with this lazy Susan, right? He pushed all the vegetables to the side, to the outside, and he pushed all the meat into the middle, like the pork buns and stuff. And he was spinning it round and I was watching him and pork bun would come round very quickly and he'd take it and the vegetables would take ages to make this sort of loop round because he wanted to avoid those. And he got me thinking, right? He got me thinking that everything we do is an empirical life cycle, right? Everything we do is an empirical feedback loop. Plan, do, check, act, build, measure, learn. What changes is the speed of experiments that go round that empirical feedback loop, right? So we came up with this. It's all one empirical feedback loop, right? It's all, we all measure stuff, we all learn about something, we'll build stuff. What changes is the experiment that we run? And that sometimes those experiments take ages to get round that loop, right? Sometimes they're really quick. If we talk to a customer, that's a really quick experiment, right? And we can probably go to a talk to a customer very quickly and get some food that very quickly. But we built some interview questions. We've learnt from it and we cycle round again. If we build some software, that's a really big, expensive experiment, right? But sometimes it's a thing we need. But it's going to take much longer to get round that feedback loop. So this form, the basis of our playbook, right? This form, the basis of how we approach product development, right? So we start, every team starts with a mission. It starts with a set of metrics. What are we going to do? How are we going to achieve these outcomes? How are we going to measure these outcomes? And then off they go and they start understanding where they are now and they start running these experiments, right? And typically we run one of these four experiments down the bottom left and we'll come to those in a moment and what we call categories of work. But basically this form, the basis of our playbook, really a high level. And then we don't really care if you're doing Scrum, Camman, Mashup of the Two, XP, whatever, right? But this is the high level thing and everyone could get behind this, right? This wasn't, this was agnostic of product management, UX, everyone could get behind this. The second thing I learnt was acknowledge the mindsets, right? Acknowledge and mash up the mindsets. I love this, it's a great book from a chap called Johnny Schneider. Check it out if you get a chance and this is a slide from him. And he talks about mashing this stuff up. There's a reason that I'm called a practices coach at B&Z, not an agile coach. I'm a magpie, right? Nick can choose things from anywhere, from UX, from design thinking, from human senate design, from lean, from agile. Nick can choose whatever works and apply it to the context of the environment. I'm not precious about that. But what is important for us is those high things like getting that value out to that customer in a disciplined way. So this again comes back to that language thing. As soon as we start mashing up those mindsets and I'm not focusing on a framework, I'm not setting up Mr Shaltypants coming out with Sprint and all these language that no one understands. We're focusing on why we do stuff not what we do. I come over who came up with this slide originally. I've tried to search it on the internet to see who came up with the original. I haven't found the original source, so let me know if you've found it. But I love this. Caution, agile has no brain. Please bring your own, right? Common sense, bring your own brain. Agile's just a framework. It doesn't give all those different steps. It gives a high-level framework. You bring your brain. You are the experts. You work out what you need to do. I think we must never forget that. Never forget that. You choose, right? This is all about you bringing your expertise, you bringing your brain to agile and you picking and choosing what the right practices are what the right things are to do. Drop that ego, drop that sort of practices. Drop that real how you do stuff and focus on the why you do stuff and let people be experts and be their conscience. The third thing was about keep it draft. Keep it draft. This really invites that collaboration and this really invites people to be experts. The key thing I did here was send it out like this with a big old draft on it. Loads of questions, loads of highlighted yellow text like how are we going to do this or how will that work? I send it to the UX researchers, and invited their feedback. That's how it became our playbook and everyone could buy into it. Your note on here, there's no author because everybody's the author on it. It's our playbook and it's an evolving thing as we go through. One of the things that really kicked in for us was about experiments. We talk about minimum viable products and that sort of segmentation between customer and the product itself. One of the things that we built on top of that sort of empirical framework was this idea of experiments and what these experiments meant. Experiments were important for us because again we wanted experts to be experts, we didn't want to water people down and we thought when we run an experiment we want to put some criteria in there, we want to make sure that it's guided by a specialist. Guided to be an expert who can be seen as being the expert but they're involving the wider product team and that wider team feels involved, they feel that they have a chance to get involved with that piece of work, that experiment. The other thing is whatever experiment is it's tracked on the team board, people know what's going on, they can ask questions, they can see the status of it and see the progress. The final thing is it's time box. Time box is really important. If we don't know what the final acceptance criteria of something is let's just bound it by time. At the end of that time we can review whether we want to run another experiment or what. These are the four experiments that we came up with. Fundamentally we thought everything we do falls into one of these four buckets. If I go to a team and they're on a piece of work I will say to them what's your next experiment? Are you going to talk to a customer? Are you going to gather some data? Are you going to prototype something? When I first put these in I had numbers on them. One, two, three, four, because I thought it would look kind of cool. Then one of the senior product managers came over and he said, can we take the numbers out? I went, yep, sure, why? He said, because we don't have to do them in that order, right? I said, no, of course not. He said we could prototype something and then we could go off and gather some data and then we could prototype something again. I said, of course you could. So we took the numbers out. Then he said, put it in production. Could we maybe change that subheading on there to make a change? Because we don't always put software into production, right? Sometimes we just make a business process change. Yeah, of course we do, right? So we changed it. Again, we got that feedback. But fundamentally this really helped us because I remember working with a team and they were working on something called self-service, right? I spoke to them and I said, self-service sounds great. What are you doing? We're vying up what experiments to run next. I said, that's great. What are you doing? And they said, well, we don't know whether it's this opportunity or this opportunity. And I said to them, well, what are your metrics? And they looked at their metrics and they said, well, we want to reduce calls to help desk, right? So I said, that's great. What help desk? We've got three and they said, well, they said, a small business hub. I said, great. So what's your first experiment going to be? And they said, well, how many calls do we get to help desk each day? I said, great. How many calls do you get to help desk each day? And they said, well, we don't know what you're going to find out. Well, let's go and talk to the customer. Off they went and off they went to talk to that help desk, right? So, again, it's me acting as their conscience and putting that lean discipline in place. So when we looked at how we run these different experiments, we looked at it like this, right? We thought, you know what? If you're going to go and talk to a customer, you probably want to get some UX researchers involved, right? A customer researcher or a UX researcher. Go and talk to those people. They're the experts and get them to lead it for you, right? Or maybe it's a team design or someone. The second one would be gather some data. And we'd say, well, gather some data. You know what? If you want to gather some data, you might want to go and talk to the data analytics team, right? Or maybe you want to come up with some new data techniques and find out what's happening and put that into production. So maybe you could do that as a development team. Prototype something. Hey, you know what? Is that a paper prototype? Do you need to go and talk to a designer? Do you need to run a design sprint? Or are you thinking a technical prototype? Do you need to run a spike? Do you need to get the team to do that themselves? And the final one, put it into production, right? This is the only one thing that delivers value, and this requires a bit more overhead and a bit more quality, et cetera. But you probably want to involve risk and security. Just to let them know, because we're a bank and we have to think about risk and security. But again, it's about getting the team to think about what they need to do at a high level within these containers. Now, when we kick this off, right, very early on, the team are pretty frightened, right? And when we first put linearX in, we kind of just said to them, hey, off you go then. Go and talk to the customer. And you could see the fear in some of their eyes, right? It was like, what? You want me to go and talk to the customer? And some of them are really frightened, right? And I think the thing that we found was, we give people the opportunity to talk to the customer. But we get them to care about the customer, care about the problem. Some developers just don't want to talk to customers, right? But we give them the opportunity to do that, and we give information about what the customer is thinking. And others absolutely love it, right? We've had testers who have been there, been testing for 20 years, and they say, this is the first time I've ever talked to a customer. You know, taking notes in a session or viewing a session or something like that. But we give them the opportunity so they can care. And it's this thing here, right, for Steve Jobs. Basically saying that you want to care about your job, right? You know, we spend so much of our time at work. You've got to be happy. We're going to make sure people are happy at their work. They're doing what they need to do, but also that they care about the customer and what they need to do. They understand the why questions. The fourth thing was about saying no, right? Treat your product playbook like you do your product. And no for a product manager is one of the most important words we have, right? And it was easy for me to say, you know, I was getting these requests left, right and center for how can we just put this in the playbook and this framework here and, you know, that and the other. And if we said yes to all that stuff, we would end up with this huge monolith, right, this huge document, 1,000 pages. So we kept it very distinct, very distinct, very focused on the product development stuff that we were doing, and we cut a lot of stuff out. This stuff from Dez Trainer, right? You've got to make sure that you end up with some crazy Swiss army knife that doesn't really meet any of your needs, right? If you want to just focus on your core product and what you need. I like this one from Dez Trainer as well. Features are like house pets. And this has trade me a big kind of trading website in New Zealand. And I went in there with my little girl and we searched for free pets. Loads of stuff comes up, right, for free pets. But the problem is not getting hold of the free pet. The problem is not developing a feature. The problem is the longevity of that feature, right? It's maintaining that feature. It's feeding it. It's supporting it. It's the technical debt that comes with it, right? And the same thing with our product playbook. And the fifth and final thing was question the thinking, not the answer, right? So again, it's about me being that product conscience for the team and the product playbook being there that allows people to ask powerful questions. I don't question the answer and why people came up with stuff. We put a couple of templates in the playbook that helped us as well. One of them was based on Jeff Patton's opportunity canvas. We started off using a lean canvas which kind of worked for us but there was boxes that we didn't really use. So we got opportunity envy and we came up and we started using Jeff Patton's opportunity canvas. And then we've adapted it since to this opportunity canvas here. And that's great, right? Because typically what we found, I'm sure you've never found it, right, is you'll get someone coming along to you and going, right, I've got this solution idea, go away and size it up and tell me how much it's going to cost. Has that happened to anybody? A few people, yeah. And we found that, right? We found that with some of our teams that they felt like they were just a what to do tapping into their brains and their expertise. So we used this opportunity canvas when we kick off a piece of work. And typically what happens is if someone comes to us with a solution idea we're like, wow, that's great. Great, you've got that solution idea. Right, okay. Can we just zip back up and we do this in a workshop, right, and we put this on the wall. What's the problem we're looking to address or opportunity? I've got to keep saying opportunity because we don't just have problems. We've got to be able to address and we go through this and assumptions start to come out. This is a living document, right? So we don't get it all at once. And then we come around to this thing called budget, right? And budget's really interesting because what we try and do is change the conversation. Rather than being how much is this thing I want, going to be, we turn around and say, right, just so we know when we start coming up with our solution ideas, how much is that problem we're solving? How much is that opportunity worth realising? Is it a $10,000 problem? Is it a $100,000 problem? A $10 million? And depending on how much that problem is worth solving, depending on how much that opportunity is worth realising, then we can cut our cloth accordingly, right? Then we can start finding out and then we can start running our experiments to make sure that we're doing the right thing within the constraints of the budgets that we're given. But we get that whole team inclusion. The other thing we do as well is all about discipline steps. Does anyone remember this game? Mastermind? Yeah, a couple. I remember it as a kid growing up, my parents had a battered version in the games cupboard, right? And what you basically did with Mastermind is you'd have two plays, right? One play at the end would put six different of these coloured pins in and then the idea of the other person had to put these pins in at the other end and they had to try and guess what you'd got here, right? And they'd do it progressively through these discipline steps and you would tell them whether they've got a colour right but in the wrong place or you've got a colour right and it's in the right place. Now, often when we approach product development, right? We just randomly chuck stuff out there and we hope that it's right. But the thing what we want to do with product development is have these discipline steps and these discipline steps should be all about right. Put something out there, right? Run an experiment. Is this looking all right? Do we think we're making progress towards what the customer wants and what the customer needs? And what we've tried to do is introduce this through a carter type approach what we call a carter or a disciplined approach. And again, this comes back to us being lean and taking, mashing up all this stuff. We've taken lean and we've introduced some product discipline steps in this carter and we call this a product experiment map, right? And again, coming back to that experiment that team ran with talking to that small business hub customer help desk. So their first experiment was, you know, they would go away and talk to that small business hub and their experiment was they get 300 calls a day, right? Bang, right, we've learnt we get 300 calls a day. What's our next experiment? Well, our next experiment should be what's the number one call, right? What are the top calls? We have all sort of gather some data around that. So they went off and they go and gather some data and they find out that the number one call is change of address. Change of address, right? So the next experiment would be if we were to put change of address online would that reduce calls and is that something the customer wants? Right, bang, off we go and run that experiment. So again, this puts that discipline into the process that we have but it doesn't let us question the answers. It questions the thought process, right? So what's your next experiment? So how do you know who have you talked to? So again, it gives that discipline. And it allows us as coaches, it allows the tribal leadership team, it allows product managers from other areas to ask these powerful questions. What assumptions have you made, right? Who's going to lead the experiment and what's your next experiment going to be? How do you know you're being successful? What's your measures of success and who's your customer? So this is a question which focus on why we do stuff and not what we do. And this allows us to govern the outcomes rather than just the outputs. This one as well, I love this Hitchens razor. Again we talk about assumptions rather than just taking requirements as wrote. What can be asserted without evidence can be dismissed without evidence. And again that assumption is that a powerful word. And we validate or invalidate that through that discipline and through these small steps. So where are we now, right? So our product development playbook it's become a field guide to running product experiments and delivering value to B&Z customers and users. It's still small, it comes in at something like 27 pages I think. And you know, look, people are using it. They're using it across the bank. It's started out as a digital thing. It's spread across the bank as a whole. We've got product managers, financial product managers are using it in lending and other areas as well. It isn't sort of focused on one sort of area in the frameworks and technology and this stuff. And here's someone here. It brings together all the best parts and practices and it's easy to understand Ray providing a framework that our team has adopted and then adapted. So again it's high enough level but it gives them the discipline that then they can start to think about how they apply it. And this is the team that's not my effort it's everyone's effort. It's all these people. We continue to evolve as we go through and it has become the banks as well. 10 minutes. That's me. Thank you very much.