Imposing Agile with Coherence, Constraints and Curiosity - Karl Scotland at FlowCon France 2019 -E05

[00:00:00] Good afternoon. thank you for coming to my session this afternoon. I'm Carl Scotland. I, um, so I work for a company called Tech Systems Global Services, and I lead the Agile Transformation Services practice there. Um, so based out of the UK. Um, we have a lot of coaches in the US and we're we're starting to build up a presence in, uh, in Europe as well. So, I'm going to talk about how to impose Agile on your organizations. That's what everybody wants to know, isn't it? How do you go in and make people agile? Um, so it's a bit like a an operating system, which is just a it's just an Agile operating system, so we can get our stack of floppy disks and install it and, uh, it'll work out fine, so it'll go something like this. Oh, not like that.
[00:00:50] Let's try it on here.
[00:01:40] So I'm guessing from the laughter in the room, that's not an unfamiliar scenario.
[00:01:50] So I just put I'm not going to take credit for this. So Jason Little put this together, put it on YouTube and gave me gave me permission. But, um, yeah, I think it's a really nice kind of way of, uh, expressing how most people have experienced Agile transformations at some point. And it's I I'm not picking on Spotify or Jason's not picking on Spotify. Um, it's just that that's what happens in organizations with with Spotify, safe, any sort of framework. Um, so what do we do instead? Um, so this this kind of visualizes this in in the form of a Wardley map. Um, so I'm not going to try and teach Wardley mapping if you're not familiar with it. Essentially, it kind of maps out the visibility of components, um, on the vertical axis and the the evolution and maturity and ambiguity of components on the horizontal axis. And what you're doing when you're copying Spotify up here, if I try and get this to work, uh, is the highly visible, highly evolved bits of the process, so it's the bits you can see and you can take and the bits that are easy to copy. And what you miss is the bits down here, is the the bits that are still emerging and and evolving and that you can't really see. So you're only kind of copying a bit of it, which is one of the reasons it doesn't work. So this talk, a bit of background to this talk, where it comes from. This is, um, a reaction to the idea of imposing Agile. Um, and and actually it's a reaction to the reaction, which I'll explain a little bit. So this is a quote from Daniel Mezik. Uh, nearly 100% tolerance of the coercive imposing of Agile practices on teams is a cultural norm of the Agile industry. This tolerance of force is profoundly disrespectful of people, this is the most pressing issue of our time. There's some fairly strong words, and I think they're fairly justified as well. Um, I think there are a lot of coercive imposing of Agile practices, um, sometimes called the Agile industrial complex. And the response to this, or the way that Daniel and a number of people are suggesting we we approach this is rather than imposing Agile practices, we use engagement models. Um, and and Daniel describes an engagement model as any pattern or set of patterns reducible to practice, which result in more employee engagement during the implementation of an organizational change initiative. So instead of just imposing and enforcing and coercing people to work in a way that maybe they don't want to work, maybe isn't the right way to work, or they don't quite understand why they've asked to be working in a certain way, we engage them in the change process and we make them part of the decision making. Um, and that's great. That all sounds good. My challenge with this, and this is why, you know, this is my reaction to the engagement model as a reaction. Is is that I think even if you're choosing to use an engagement model, um, such as, uh, you know, agenda shift or open space agility, um, or, you know, clean language could be thought of as an engagement model, I think you're still imposing that engagement model on the organization. You're just imposing something at a higher level up. Um, which is not to say engagement models are a bad thing. So the the the the paradox there, um, and this is kind of what really led to this talk, is is what makes imposition good and healthy and work well. And what makes imposition bad and negative and and lead to failure? So this talk is just basically a bunch of ideas I kind of, I was kind of pondering on this question, and I ended up coming up with the three Cs. So I first did this talk in in the US, so I had to kind of get some cultural, American cultural references in there. So the Sesame Street, this talk is about do you buy the letter C and the number three. So I'm going to talk about three Cs. Um, and there's a number of other models in there with three Cs in them as well, and I'll leave that up to you to spot them. Um, but the ones I'm going to talk about, the first one is coherence.
[00:05:35] So, when we go through an Agile transformation, we want it to be coherent. So what we're imposing is a form of coherence. Um, and I'm using the definition of coherence here, the quality of forming a unified whole. So, let's try and drill into that a little bit. Going way back into the the 19th century, um, the predominant prevailing wisdom, actually, I think it's still the prevailing wisdom, is we talk about alignment and autonomy. Uh, and and the thinking was you could have one or the other, but you couldn't have both. Because if you had high alignment, you achieved high alignment by giving people detailed instructions, giving people a plan and saying execute this plan, uh, and therefore they have no autonomy. Or we can go for high autonomy by saying, go off and do what you like. You have complete freedom of choice to take whatever action you need, and therefore people are going to go off and do random things, and therefore you're going to get no alignment. So that was the thinking. How what do you do? Do you want alignment or autonomy? They're kind of two ends of this scale. Um, hmm, interesting. So then a guy called Helmut von Moltke, who's chief of staff of the Prussian Army in the late 19th century, said, you know what? Actually, they're not two ends of a scale. They're completely two different, completely different axes, and you can have high alignment and you can have high autonomy. So you achieve high alignment by making clear what your intent is. You don't give people a detailed plan, but you give people the intent of what you'd like them to do, and once people understand the intent of what you'd like them to do, then they can have high autonomy to take the right action to meet the intent. So, clarity of intent means that people can take action. Um, Steven wrote about this in a great book, The Art of Action, and I have a slide at the end which has all the book references in it. Um, and he he describes this as directed opportunism. So this is this is the model he describes, and he says there are three basic building blocks. Uh, there's the outcomes you want to achieve, then you make plans to achieve those outcomes. Then people take action based on those plans, and those actions will lead to outcomes. And in the perfect world, the actions that people take lead to the outcomes that you originally wanted and intended. But he describes these three gaps, the knowledge gap, the alignment gap, and the effects gap. The knowledge gap, the difference between what we'd like to know and what we actually know. So when we're making plans, we don't necessarily know everything, so we can't make perfect plans. So that's the knowledge gap. And then we have plans and we want people to put them into action. The alignment gap is the difference between what people do and what people we want people to do. So people don't necessarily follow the plan exactly as we intended, so that's the alignment gap. And then you've got the effects gap, so when people take action, the results aren't necessarily what we expect them to be. So things don't happen in quite the same way as we we want them to. So the effect gap, the difference between what we expect our actions to achieve and what they actually achieve. So to get the the perfect outcome, we need to figure out how to close these gaps. So the typical way, the one that most people are probably familiar is, is basically go for more and more detail. So we close the knowledge gap by getting more detailed information. Let's do more and more research, more and more up front work. And then we close the alignment gap by making our plans really, really detailed, so lots and lots of micromanagement. Um, and then the effects gap, um, so sorry, lots of detailed plans and then we manage the actions by micromanaging people. So lots of detail, we have lots and lots of detail controls in place. Tends not to work out so well. And what says is instead of going for more detail, we should go for less detail. So we limit direction. So when we're closing the knowledge gap, we limit direction to defining and communicating the intent. So we don't tell, we don't do lots and lots of research about what we need to do to make the plan. We just say, here's our intent. And then we allow people to come up with their own plans and come up with their own actions, um, themselves. And then he talks about this idea of back briefing. So we people kind of say, this is what we're going to do to meet the intent, um, and this is how we think what we're going to do meets the intent. So you people brief, they brief back to you or back brief so you can make sure that you have got that alignment in there. And then once people start taking action, you give people the freedom to to change their mind and adjust, um, and and, you know, that inspect and adapt process so that once you're taking action, you can start figuring out is what we're actually doing helping us meet the intent, are we achieving the outcomes that we first desired? So actually do less less detail and give people more freedom. Um, Elizabeth mentioned this earlier, another very similar model that David Marco talks about, um, in his book Turn the Ship Around. He talks about giving control to people. So he, as the commander of a nuclear submarine, could do a couple of things. One, he could just give instructions, tell people what to do. Uh, it turned out he didn't know much about the submarine he was posted on. Um, he realized that the people, his crew, knew more about the submarine than he did, so he wanted to give control to the people, which was a strange thing to do as a, um, in the Navy at that time. Um, and he realized that in order to give control to people, one, he needed to develop people's competence, so people needed to be good enough at their jobs, and he needed to, uh, make sure people knew the clarity of intent. So people could describe why they're doing something, they had the ability to do it, and therefore you could give the control down. So there's two very similar models there. Um, so hopefully under your seat, you should have some 3D glasses. No, do the do the the organizers not put them out? Okay, you're going to have to squint then. So I find this is useful. If you just do this with your eyes. Okay. All right. Excellent. There's always some people that actually do that as well. All right. So imagine this is this is three dimensional. Um, I basically bash those two models together. So you've got, um, the autonomy and the alignment in pieces, and I've added a third element, ability, which maps on to autonomy, giving control, uh, alignment, giving people clarity, and then the ability, uh, giving people competence, competence. So another three Cs and three A's in there as well. And then actually, you don't just kind of do that. You don't kind of go, ah, you're now, you're now competent, and you now have clarity, so I can give you control. You kind of have to, this is the squint bit, you kind of want people to kind of go on a bit of a journey. So imagine that arrow is coming out of the screen at you. So you slowly kind of give people more alignment and give people and make sure people have the right ability, and then the more alignment people have, and the more ability people have, the more autonomy you can give them. So this is this is, um, that that alignment and that people have the ability to to meet the intent is is what I think of as coherence.
[00:12:42] Um, there's another model I'm going to talk about, uh, and I'm going to reference a few times, called the Four Disciplines of Execution. And I think Elizabeth referred to this in the keynote this morning as well. Um, Four Disciplines of Execution has four disciplines. It's it's quite aptly named. Um, discipline one, um, is is a discipline of focus, focusing on widely important goals. Um, so what what 4DX says, for disciplines of execution is, is make sure you know what your goal is, and make sure it's a really important goal. Um, and I just kind of pulled out a few quotes that I I like. Um, just to help kind of what do they mean by a widely important goal? So if every the area of our operation remained at its current level of performance, what is the one area where change would have the greatest impact? So you really kind of get people to think about focusing on one area of the business where you're going to have impact, that's kind of your intent really. Um, and then they add some things around, you know, being able to measure it. So if it's that widely important, surely you should be able to tell if you've achieved it or not. And if you can tell whether you've achieved it or not, you can measure it, and they have this nice little formula from X to Y by then. So X, this is where we are now, Y, where we'd like to get to. So what's our intent? Um, and then by when, let's put some time scales on this, are we talking a week, a month, a year, five years? Um, so they talk about this wildly important goal as a wig, WIG, um, and it being worthy and winnable. So, um, it should be ambitious enough that it's worth spending time on, um, but not so ambitious that people are kind of go, what's the point? We'll never do that. So I kind of like that as a a another simple way of of gaining coherence. Just make sure people know what your widely important goal is. So in Agile transformations, that really means being able to answer these questions. Do we understand what the intent is of our Agile transformation? And a lot of places I go in, um, the Agile transformation is just because somebody's decided that we're going to go Agile. Somebody's got some budget to spend or somebody went to a conference and it thought it sounded quite cool. But actually, how's it going to intend in, uh, improve the business? So what's the widely important goal that we want to achieve and which our Agile transformation is going to help us achieve? So, I'm thinking about these two questions. If you can answer those two questions about your Agile transformation, you're probably going to start getting some sense of coherence. So that's the first C.
[00:15:04] Um, next one is constraints. So, uh, a limitation or restriction. So by imposing some form of constraint, um, we can start, uh, allowing people to make their own decisions about Agile transformations. Now, when I talk about constraints, I'm talking about a specific form of constraints. So I think a few people in the talk earlier put their hands up if you, uh, people know Kinevin. So yeah, kind of most people know, I'm not going to spend a huge amount of time on it. Uh, you've got the, uh, the, the five quadrants. Sorry, the five domains. Um, so obvious, which is where everybody knows. Uh, this is out of date already because Dave Snowden now refers to this as clear.
[00:15:45] Everybody knows how to how to approach something, that we can talk about best practice. Um, complicated is where experts know, and we can have the notion of good practice.
[00:15:55] Um, novel practice, uh, where we just chaos, we just need to to act and do something. Um, and then complex, which I kind of like the idea is that's where experts disagree, there's no one right answer. Um, the the relevance for constraints here is you'll see they'll talk about, um, fixed constraints, governing constraints, and enabling constraints. So, when I'm talking about imposing constraints, I'm really talking about imposing enabling constraints.
[00:16:22] Because most of the time we're dealing with complex scenarios, there's no one right way. You can't just we don't, in fact, what people try to do with Agile transformation is get an expert in who can say, here's a good practice for scaling Agile in your organization, just do it this way. Um, even worse, some people think it's best practice and obvious, but actually most organizations are different. It's a complex problem.
[00:16:42] So the idea of enabling constraints, um, really kind of goes back to Alicia Juarez Juarez. Um, so I've not found a good quote directly from her, but this is paraphrasing her from the blog at the bottom. Um, it's her her kind of way of the way she thinks about enabling constraints. Governing constraints hinder access to do something or only allow them to do it in a certain way. Enabling constraints make it possible for access to do something that would not possibly not be possible otherwise. So this idea of governing constraints narrowing you down to to one way, enabling constraints opening up possibilities. So this is the way I think about it. Governing constraints are what guide what should not be, so they're kind of narrowing you down and close close options down.
[00:17:22] So you can't do all of this stuff, you have to do it that way. Whereas enabling constraints open up possibilities. Then it's not so open that anything goes. The constraints kind of bound those possibilities, but there's an open set of possibilities in there. Um, that the metaphor that Alicia uses is the skeleton and the endo and the exo skeleton.
[00:17:42] So the exoskeleton, um, it has external skeleton, think like crabs, shellfish, where the shell is on the outside. They constrain growth, so there's only certain kind of amount of growth that can happen versus something like a starfish. And actually, as humans, we have the endoskeleton, an internal skeleton, there's much more freedom and creativity to growth to grow on top of the skeleton. So you get a much wider variety of shapes and sizes. That's kind of the the metaphor there. So we're looking at a governing constraints so that people are not just doing any old thing, but there's there's enough of a constraint in there, but we're allowing people to do their own thinking. So, quick test then based on that. This is, uh, I run marathons, well, I run one marathon a year. I don't want to say I kind of run hundreds of them. Um, this is my training plan from last year. Um, and you can see it has, um, weeks and dates and then, you know, the different types of runs I needed to do, and you can see I kind of start ticking them off and when they're when they're doing. So, um, hands up if you think this is a a enabling constraint.
[00:18:53] Okay. All right. Hands up. No, tell you the hand, if you think this is a governing constraint. All right. More people think this is a governing. So, trick question.
[00:19:02] Um, I think it depends on how you use it, not on the thing itself. So, I could use this as a governing constraint, where I literally follow it to the letter, and I do those specific runs on those specific dates, and basically I don't think about it. I just it's just a set of instructions. Or, and this is what I actually do, I use it as a governing constraint where it kind of guides me. I know roughly I should be doing three runs a week, and those runs should be around about these sort of distances or paces. Um, and, you know, ideally maybe they're in this order. But actually, what I tend to do is, you know, I'll I'll I'll be there or thereabouts. I'll make sure I'm doing roughly the right distance and roughly the right pace. And some weeks, um, I think, you know, this week down here, I didn't do this one. I just couldn't fit it in. more people think this is governing. So, trick question. Um, I think it depends on how you use it, not on the thing itself. So, I could use this as a governing constraint, where I literally follow it to the letter and I do those specific runs on those specific dates and basically I don't think about it. I just, it's just a set of instructions. Or, and this is what I actually do, I use it as a governing constraint where it kind of guides me. I know roughly I should be doing three runs a week, and those runs should be around about these sort of distances or paces. Um, and, you know, ideally maybe they're in this order. But actually what I tend to do is, you know, I'll, I'll I'll be there or thereabouts. I'll make sure I'm doing roughly the right distance and roughly the right pace. And some weeks, um, you know, some examples down here, I think, you know, this week down here, I didn't do this one. I just couldn't fit it in. So, I kind of look at that and I use it as a guide to help me work out what type of running I should be doing in what weeks. But I don't follow it originally. So for me, I think that that is an enabling constraint. So it's more about how you use it, and it's really an enabling constraint helps people think for themselves. We hire highly skilled people with lots of experience, um, and lots of potential to do great work. Uh, so let's, you know, people aren't our greatest assets. People are our any asset. Let's make sure we're using those people to the best of their abilities, rather than saying, great, you're really skilled, you're really experienced, we're paying you lots of money. Now, don't think, just follow instructions. That's what happens on a lot of agile transformations. So enabling constraint is is about really helping people think for themselves, uh, and and make their own decisions. Rather than just following instructions.
[00:20:39] So making decisions then brings us around to strategy. And one way I think of enabling constraints is, is strategy is an enabling constraint. Because strategy, well, good strategy doesn't tell you what to do. Um, bad strategy does sometimes because it's more of a plan. But good strategy, um, this is this quote from Mintzberg, a pattern in a stream of decisions. A good strategy just helps you make decisions, and if it's if it's a clear strategy, you should start spotting that pattern of decisions over time. Um, actually, um, Mintzberg kind of changed his language on this and talked about it in in action decisions, so stream of actions. Um, because we can make a decision and not act up, not act on it. Um, so we make decisions, we act on them and then if those actions are coherent over time, we should spot that pattern and that pattern should be effectively our strategy.
[00:21:28] Richard Rumelt talks about strategy as well. Um, and again, he talks about diagnosis, understanding what the problem is. And that problem, solving that problem effectively is is our wildly important goal. So that starts giving us some coherence. Then we put some coherent policies in place. Those coherent policies are the enabling constraints within which we can then take coherent action. So we can make decisions, but that coherent action, we want the right people to be making those decisions and what the action should be.
[00:22:01] Um, there's a really nice exercise that I, that I use when I was talking to people about strategy and enabling constraints and making decisions, um, and it's this one called even over statements.
[00:22:10] Um, so, an even over statement has has two positive outcomes.
[00:22:14] Positive outcome A and positive outcome B. Both positive, both are desirable. But it forces people to think about the tradeoffs between A and B, and which one they would choose given that tradeoff. So you end up saying, I we choose to achieve or to focus on positive outcome A even over positive outcome B. So the even over bit is important, it emphasizes the the the difficult decision. So it's not, hey, we're going to do positive thing A over negative thing B. It's positive even over positive. Um, so these are basic instructions that you can you can go and look on the website. But yeah, work with people and think about all the good things we want to achieve. What are all the all the positive outcomes and then let's start working around what are our tradeoffs? If we can articulate our tradeoffs like this, we've probably got a good sense of what our strategy is. Um, I worked with one organization, um, uh, started this year. Um, and what we found is that the the the the main outcomes they were talking about were around um, building their product and selling services and training around their product.
[00:23:27] And we went off and we were doing did this exercise. And basically half of the people came back and said, uh, we value product even over services. And the other half of the people said, uh, we, we want to do services even over product. So we highlighted a kind of a a misunderstanding and a mismatch in there. And we had quite a long, painful, but productive conversation where, um, people talked about why they, why some people thought we should be doing product, why some people thought we should be doing services. And eventually they came to some consensus around around what they meant. But it was, you know, it was a really important conversation that needed to be had. What's interesting, and you maybe have spotted or thinking already, is that that idea of, um, valuing one thing over another is kind of like something something you might have seen before. Um, so the Agile Manifesto was kind of like that. So this got me thinking, what if we thought about the Agile Manifesto not as a set of value statements, but as a set of strategy statements? And we kind of converted these and we took to them and we made them instead of just we value left over right, we say, uh, and and right generally is interpreted as negative. So it's normally we value good over bad. That's how a lot of people interpret it and I wish it wasn't intended that way. But let's really emphasize the positive nature of both sides and we turn them into even over statements. Um, so the first thing is we say we're uncovering better ways of developing software by doing it and helping others do it, through this week we have come to choose, because it's about choice. And then, so, I disclaimer, I did this very, very quickly. I've not got back and revisited, it's not perfect. I am not suggesting this is a rewrite, this isn't Agile Manifesto version 5.0. Um, this is just a bit of a thought experiment and I'll actually do this with organizations. Just get them to write your own Agile Manifesto. So, individuals and interactions.
[00:25:19] Even over standards and consistency. So why do we want process and tools? Process and tools can be positive. And there maybe the positive aspect of them is about having some standards and consistency around the organization. Uh, working software even over recorded decisions and knowledge. Customer collaboration even over making commitments. And responding to change even over meeting commitments.
[00:25:46] Not perfect, I know, but I think it's I find it an interesting thing as around. The interesting thing that is that I find interesting here is is taking that right hand side and saying, what's positive about this? And what is the tradeoff we're making? And actually, um, you know, particularly the last one is, you know, responding to change even over meeting commitments. Don't we want to meet commitments?
[00:26:09] I'd I'd recommend trying this with with your teams just as a just as a thought exercise.
[00:26:15] Um, the other way of thinking about enabling constraints, uh, is measures. So, so putting some measures in place where again, we're not saying what you need to do to to to generate those metrics or to, you know, to move the needle on those metrics, but we're saying these are the things we're focusing on. Um, and Larry Macheroni, um, so I worked with Larry when I was at Rally.
[00:26:38] Um, he he came up with this idea of ODM, outcome decision insight measures. So he he says, when you're trying to come up with a measure, the first thing you want to do is think about what outcome you're trying to achieve. Once you understand your outcome, and those outcomes could be the same outcomes you have in your your even over statements. So we we want to achieve this positive outcome.
[00:27:00] What decisions do we need to make in order to achieve that outcome?
[00:27:07] And then, to make those decisions, what insights do we need to have in order to make those decisions?
[00:27:13] And then what measures do we need to have in order to generate the insights? So you work back from outcomes to decisions to insights to measures. And then hopefully, once you've got the measures, those those metrics, that data will give you the insights. The insights will allow you to make good decisions, those those good decisions will lead to the positive outcomes. So rather than just kind of going in and again, diving straight into the the measures and the popular measures and copying measures and cargo counting measures. Let's make sure we again, we've got coherence, we understand how those measures are going to help us to achieve our outcomes. So a simple example would be, we want to, uh, we want to be more responsive. So, you know, we want to, we want to be able to deliver our software more quickly.
[00:27:58] Um, so the decisions might be, well, we need to decide which work we should be focusing on to get it finished. Okay, so what insights do we need to get to decide what work we should be focusing on? Well, the insight is is what work have we started and and how old is that work? So then the measure is is about feature aging or, you know, work aging. So how long has a piece of work been in progress since it was started? Now, most people when they talk about responsiveness will will talk about cycle time or lead time. But actually a measure of cycle time, you only get when the thing is finished and it's too late then. To really make good decisions and get that outcome of responsiveness, you need to know how old a piece of work is before it's finished. So that helps you kind of focus on one feature aging and then highlighting which work is getting too old and that's the work we should be really focusing on trying to get finished.
[00:28:50] So back to four disciplines of execution, um, discipline two, act on lead measures. So this is where the measures become enabling constraints. Um, and I like that definition of lead measure, um, in that it's got these two basic characteristics. So it's predictive, so we predict that
[00:29:07] um, measuring and and improving a measure will lead us to achieving our wildly important goal. So it's predictive, and it's something that we can influence, so it's something that we can act on directly. Whereas generally, uh, the wildly important goal, it's a lagging measure, it's some kind of fairly big and hairy and it's it's, you know, we can't just work on it. We can't, we can't just make it happen. So we work on lead measures, um, again, it's a measure. It doesn't tell us what we should be doing, it's just been saying, do something to move this measure. Um, we can act on act in order to move that lead measure. And then discipline three is about visualizing those measures. So having a compelling scoreboard. Um, so telling the team where you are, where you should be, and giving you that information, allowing generating those insights so that you can solve more problems, make good decisions, so that you can act on the lead measures. So that scoreboard and those lead measures again are a way of thinking about enabling constraints. Not telling you what to do, but giving you some some boundaries within which to work.
[00:30:13] So the third, the third C then. Actually, no, sorry, before we move on. Um, so some questions about constraints.
[00:30:22] So if you've got good constraints, you should be able to say what your guiding policies are, what the hard choices and tradeoffs you're making within your transformation. Um, hopefully you've got a good set of leading measures and some some ways of indicating, uh, where you're making progress and and things you can influence and and work on directly. Um, which gives you evidence, so evidence that you're winning. So if you've got this from a very a metrics way of looking at constraints, um, again, not telling you what to do, but just saying focus on these things and hopefully we'll, we'll make progress and we'll achieve our wildly important goal.
[00:31:01] So the last one then, uh, curiosity. So the curiosity is the strong desire to know or learn something. Um, now you can't impose curiosity, but I think you can do things which, um, encourage curiosity. So, quick test to start off with. Um, so if you've if you've done this before, um, kind of keep quiet. Um, you can do this online, but I'm going to do a kind of an interactive version. So this, uh, there's a puzzle here. There's a sequence of three numbers, uh, and that sequence of three numbers, and I'm going to give you an example, 248 conforms to a rule.
[00:31:38] And your your job is to try and figure out what that rule is. So what I'm going to going to ask you to do, just have a think, maybe talk to the person next to you. Um, in order to find out what the rule is, you're going to give me a different sequence of three numbers, uh, and I'll tell you whether that conforms to the rule or not.
[00:31:56] So, just like talk to the person to the person next to you, talk about what different sequence of three numbers would you use to try and work out what the rule is.
[00:32:07] I'll give you a minute just to think about that and talk to the person next to you.
[00:32:24] Right, somebody want to shout out a sequence of three numbers that you want to test and I'll tell you whether it conforms to the rule or not.
[00:32:35] I heard something?
[00:32:38] Three?
[00:32:41] What, three threes?
[00:32:47] 39, sorry, I missed the last one. 3981. Uh, yes, that does conform to the rule.
[00:32:56] All right, another one.
[00:32:59] Sorry? 1, 2, 3? Yep, that conforms to the rule.
[00:33:04] All right, so, I'm only going to give you two guesses, just in the interest of time. Who thinks they know what the rule is at the moment?
[00:33:12] People are thoroughly confused. Anybody want to guess the rule?
[00:33:21] All right. The rule is, actually. So, so most people guess that the rule is either the number doubles, so two doubles to four, four doubles to eight.
[00:33:32] Um, or, um, it's two um, two times two is four, and then four times two is eight. Sorry.
[00:33:43] That's the same thing, isn't it? There's two variations of this. But people kind of, so what people do is they look at it, they guess what the rule is, um, and then you, you, um, give a sequence of three numbers which tries to confirm your hypothesis.
[00:33:57] So, people will kind of go three, six, 12. Yeah, confirms to the rule.
[00:34:04] Okay, what about, uh, 4, 8, 16? Great, confirms to the rule.
[00:34:10] The rule is, the numbers simply increase in magnitude. It's a really simple rule. So 1, 2, 3.
[00:34:19] Most people won't get there because they're just confirming confirming their hypothesis. To really generate information, you have to disconfirm your hypothesis. So this is this is information theory.
[00:34:31] Um, probably failure, if your probability of failure is up here, so you're never going to fail, you're not not going to generate new information. So if you're always correct, so if you're always giving sequences of numbers that conform to your hypothesis and I'm saying that's great, you're not actually learning anything. Equally, if you're always getting wrong answers and your probability of failure is probably still learning nothing as well. Um, you want to be somewhere in the middle, so you get the most information when your probability of failure is around about 50%. Now, I kind of caveat this two ways. One is, um, when I get on an airplane, I don't want to be generating any new information. So I'm not saying fail 50% of the time regardless of context. Okay, context is important here. Um, and also you probably want you want failure to be safe as well. So, um, you want kind of small failures that you can learn from without risking people's jobs, lives, businesses, those sorts of things. Um,
[00:35:27] to do that, um, you quite often you also want to be intentional about failing. And what you want to learn, it's not just about doing stuff, failing and kind of going, ah, okay, we failed. To to really generate information, you need to be intentional about it. Um, people familiar with this game, Mastermind? Okay. So I'm running a workshop directly after this with Matt Phillips and we're going to we're going to do this exercise. But I'll explain it briefly, um, because I did this with my daughter at Christmas. So this Christmas, this last year, um, she kind of my daughter's um 14 now. She kind of pulled out a really old copy of of Mastermind and like, can we play this? And I was like, it was Christmas afternoon. I didn't want to be playing Mastermind. Um, but I thought, oh, actually, I know what I can do here because because this is kind of scientific thinking.
[00:36:12] So basically, I I taught her, um, the the really easy way of solving the problem in quite a small number of steps. In such a way that it just took all the joy out of the game and she didn't want to play anymore. So,
[00:36:28] the way to do this, most people will start off with four different colors. I'm kind of thinking, if you're going to come to my workshop, I'm about to spoil this for you, so, uh, block your ears if you if you don't want to know the answer.
[00:36:42] Um, when you've got four different colors, what are you learning?
[00:36:47] You're you're learning whether one of those four colors is in there, or maybe multiple, but actually, if there's two different colors in there, you don't know which two. So actually there's there's quite a small amount of learning in there. The way I told my daughter to play this is I just do just a row of whites. Because what I want to learn is, are there any whites in the solution, and how many are there?
[00:37:09] And then once I know that, I can do, let's say there's there's there's one white. Now I know there's a white in there, I just don't know which one is white. So now I can put a white in and three blacks. And now I'm going to learn whether the white's in the right place and and maybe if there's any blacks as well. And I can slowly build up on that and it seems like it well, it is quite a slow and painful way of doing it, but actually you get to the answer very, very quickly. And as I say, it takes all the fun out of it as well. But it's that idea of being really intentional about learning something small, which actually makes you more likely to fail. Because if I'm only doing whites, I've got more chance of getting, you know, no white or black pegs on there to give me any feedback. Because if there's no whites in there, there's that you know, arguably I've failed. But I've generated quite some really useful information now because I know there are no whites in there. So, um, we'll do this later and we're going to basically we get people to write down, um, what they want to learn, what their guess is, whether they did learn it and we kind of try and do this intentionally.
[00:38:08] Um, so being open to failure and kind of running experiments is is about being curious. Um, and a lot of it is is kind of saying, I don't necessarily know the answer. So I want to learn. That's why I want to run experiments because I want to generate information because I'm recognizing that I don't have all the information.
[00:38:28] Um, there's a there's a set of questions in a way we're called clean language and I think there was a there was a I think I saw on a program this morning a session on clean coaching. Um, so Caitlyn Walker wrote a book on clean language called from contempt to curiosity. Um, and she describes contempt meaning, um, thinking that a reaction and attitude of a person or a group of people are unacceptable. So, I'm right, you're wrong, you're unacceptable. I have contempt for you. Versus being curious, just noticing how things are, wondering how they've come to be like that and wondering what we might want to happen next. So being much more open about, hey, I don't know all the answers. So I wonder why you think something differently, and, uh, I wonder what's led to you thinking that and, uh, I wonder what we can do with that information. So, clean language has has some, um, very specific, intentionally designed questions. Um, and there's there's a whole bunch of them. I'll just kind of introduce some like the three most powerful ones I think they are. So the first two are called the lazy Jedi questions. Um, and it's like, you know, the Jedi bit from uh Star Wars, like these are not the droids you're looking for, they're kind of quite magical questions that can can get some very interesting answers. So the first is just what kind of X? So X here for all these questions, X represents a word. So you're you're talking to somebody and they'll they'll answer using a particular word or phrase, and you take that word or phrase and you say, hmm, what kind of thing? What kind of X? So, you know, I'm talking about pair programming. Now I could say, oh, I'm an expert in pair programming, so I know exactly what pair programming is. Or I could think, well, okay, I have my view on what pair programming is. and there's there's a whole bunch of them. I'll just kind of introduce some like the three most powerful ones I think there are. So the first two are called the lazy Jedi questions and it's like, you know, the Jedi bit from Star Wars like these are not the droids you're looking for. They're kind of quite magical questions that can can get some very interesting answers. So the first is just what kind of X. So X here for all these questions, X represents a word. You're you're talking to somebody and they'll they'll answer using a particular word or phrase. And you take that word or phrase and you say, what kind of thing, what kind of X. So, you know, we're talking about pair programming. Now I could say, oh, I'm an expert in pair programming, so I know exactly what pair programming is, or I could think, well, okay, I have my view on what pair programming is. But maybe your experience and your view is different. So I can say, what kind of pair programming? So it's a very neutral question that can instill a wide range of answers and that's why we call it clean, so it's not biased and dirtied by my my views and my experience and my meaning and my bias. It's clean, it's open. So I can just say what kind. And then you might kind of answer it's like, well, when two people sitting together. Okay. Well then is there anything else about pair programming?
[00:40:34] So again, it's just getting people to talk about their metaphors, the way they think about things. So those are the lazy Jedi questions and I've I've there's there's French translations, I've put the URL so I'll share the slides and I've just put the French translations up here for you. And then the power switch session is, what would you like to have happen? So a lot of time people talk about problems and challenges.
[00:40:59] And this is the power switch because what we're doing is we're getting people to stop thinking around problems and challenges and to think about outcomes, positive outcomes. So what would you like to have happen is kind of going, well, what what good future, what positive future would you like to achieve? Um again, they're very thought-provoking questions if you're if you're familiar with the idea of powerful questions. I kind of think these are like really powerful questions. Um and that they're very well designed. So, let's I'm going to give you a few minutes just to try this, just to kind of break this up a bit. So again, find somebody near you, turn to the pair next to you, um and then one of you talk about this this question, the prompt at the top, so what's it like when you're working at your best? And again, this is another kind of classic clean type question, um because it's not, it's not assuming any particular answer. It's just, imagine you're working at your best, what's that like? So one of you ask the question and then the the other person, um answer the question. And then keep the conversation going, the person that's asking the question, only asks uses these three questions, what kind of, is there anything else about and what would you like to have happen? You can use them as many times as you like. You can use them in any order, but just you have to pay attention. Listen to what the other person's saying, pick up on some of their language, use their language to ask the next question. So I'll just give you a couple of minutes to find somebody, turn to the person next to you and just have a conversation, one of you asking, the other one answering about what it's like when you're working at your best.
[00:43:00] Yeah, that's good work.
[00:43:51] Okay. So that was a really kind of rapid, rapid, uh
[00:44:00] experience for you there. I'm kind of just kind of wondering, so just like hands, hands up because I know shouting out doesn't work. Does anybody have a kind of an interesting conversation? Has anybody have any interesting insights from that? I know it's kind of really short. Um, I find the the ambiguity of the questions really makes you think. And sometimes the phrasing of the questions because they're quite unusually phrased, force your brain to kind of work slightly differently because you know, they're not leading questions at all. They don't kind of don't take you down a path that makes it either easier to answer. You really have to think about it. And then as the person asking the question, you really have to pay attention and listen to what the person is saying. And by giving that person attention, again, you're having a much more powerful conversation. So I like this idea of of clean language and these questions as a way of me not assuming that I know all the answers and therefore I am I'm naturally being curious about what other people think and what other people's experiences are and then I can take those experiences and those ideas and we can we can build those into the our ways of working. So I'm not come and going in as an expert, um, I kind of go in and find out what other people think. Um, there's another way of doing this if people are familiar with Mike Burrows and his agenda shift, he has, um, an exercise that he calls 15-minute photo, so photo there is from obstacles to outcomes. And again, he's using a lot of these clean language questions to work out, and we use this within our Agile transformations of, um,
[00:45:26] working with teams to find our, first of all, find out what their obstacles are, so what are the things that are getting in their way to to being effective. And then we we use that what would you like to have happen question to flip that to an outcome. So, um, what would we, if we remove this obstacle, what would we like to have happen? And then we can explore that. And that leads us to coming up with new new ways of working, changes that we'd like to happen. Um and it's engaging. This is kind of as I said earlier, gender shift is a is a form of engagement model because people are able to come up with answers themselves just through you using some of these curious questions. Um, the other thing I like about clean language is I think it it relates to Chris Argyris and his ladder of influence. Um, so Chris Argyris was a kind of a a a ballooning organization specialist and his ladder of influence, um, says that you you start at the bottom of the ladder. So the bottom ladder, we have the whole of reality that we could observe and that exists out there. And what we do is we select a subset of that reality that we're interested in. And we interpret some meaning from that reality and from that interpretation we start making assumptions and drawing conclusions which lead to beliefs, and then we take action. And what we do is we kind of climb up that ladder really quickly. We're very quick to take action, um, and what we need to do to be curious. We need to kind of climb down the ladder. We still need to start revisiting our what beliefs led me to to make that decision to take that action. And what conclusions did I draw which led to those beliefs and what assumptions have I made which led to those conclusions? So I find clean language is a way of climbing down that ladder. Because what you're really doing is trying to find out what the other person's selected reality is and what the other person's meaning is and what assumptions they're making. And once you understand the other person's reality, the other person's meaning, assumptions, etcetera, and how they're different from my own or ours, we can start trying to do something about those differences or we start understanding each other better. Um, so just being intentional about um, just called assertive inquiry. We still, we still assert our own beliefs. And we still, you know, advocate for our own action, but we do it in a way which inquires for why other people think something differently. So we don't assume that we're right. So I like kind of that that model as well for being curious is coming down the ladder of inference.
[00:47:56] Um, in terms of lean, I kind of think of being curious as is is what lean calls catchball. So we're not just assuming we have the answer, we're kind of we we have an idea, we have a suggestion, that's the ball in catchball and it kind of gets thrown around the organization. So, you know, we throw it to somebody, they catch it, they'll reflect on it, um, give feedback on it, hopefully improve it, and then they'll pass it on to somebody else. And by by throwing ideas and including everybody in the organization in the transformation and coming out with ideas, then we're going to get consensus. So catchball, I define it as a consensus process to translate the organizational strategy into the daily tactical work.
[00:48:34] We have, we have our coherence. We have some enabling constraints within which people we want people to work. Now we can be curious with people about what things they're going to do. What do we actually do to start achieving our goals?
[00:48:50] Um, I think of it as as improvisation as well. So, um, Neil Malik is a is a, um London-based, or UK-based, uh improv. So he's he's on, he was on, whose line is it anyway, um, in the UK and was one of the founders of the Comedy Store in London. He now actually, he's he's he's he's his day job effectively, he now spends a lot of his time going around teaching improv to executives in businesses in order to help them, um, with their work. So he defines improv as these three, these four, sorry, five things longer. So listen for offers. Accept those offers, so you always accept, so if you're kind of familiar with with some of the Scrum, you'll say yes and. You never say no. So that yes and is accepting offers.
[01:39:34] Give offers in return, so that that catchball idea of improve and pass it back. Explore your assumptions, so that's that curious idea and bringing people down the ladder of inference. Um, so you explore your own assumptions, explore the other person's assumptions and then re incorporate previous ideas. Start start blending people's ideas and experiences and uh what's happened in the past and what ideas we have for the future. So this is this is rules of of improv, but we can apply that to the business. So, you know, let's not assume we have, you know, we're just not going to follow a script, we're going to improvise our way through the business. So another nice little metaphor I like. Um, and lastly on this, um, liberating structures. I think there's a session later today or tomorrow on liberating structures. Um, really I would kind of recommend going to that really nice set of exercises, which is basically is a way of including people in in decision-making processes and and getting everybody's input into into a process.
[01:40:31] Rather than just one person making those decisions. Um, a bunch of exercises in there. Um, kind of just go to the website. It has this image. They're all clickable, it takes you to a page with instructions on it. Um and it's basically, you know, when I'm running workshops with teams to to figure out what are we going to do next, um I'll quite often go to this as my my starting point for working out how I'm going to run that.
[01:40:55] So back to four disciplines then. Discipline four is is creating creating a cadence of accountability. Um and it's the cadence that's key here and it's the thing that I was I think I like most about four disciplines of execution, which I think is missing from things like OKRs. Um, where it's I don't think it's explicit enough and you end up with your outcomes, key results and it just becomes a plan and it's like go off and follow the plan. The cadence is is it's a bit like a stand-up meeting. Get together on a regular basis, daily, at least weekly, and look at your lead measures, look at your scoreboard, talk about how well you're doing toward your wildly important goal, and decide what you're going to do next. And you're allowing people to decide for themselves what they're going to do next. You're not saying, right, here's what you're going to do, here's what you're going to do, here's what you're going to do. You're saying, each person is answering this question, so what are the one or two most important things I can do this week to impact the team's performance and the scoreboard? So allowing people to take ownership and be accountable for the change itself within the kind of context of the previous three disciplines. So that's kind of this is the bit that I think ties it all together. Again, we're not assuming we know the answer, we're being curious and allowing other people to figure things out for themselves.
[01:42:09] So if you have curiosity, you should be able to answer these questions. So what hypothesis can you test? Are you testing a hypothesis or are you just doing something doing what you've been told? What do you want to learn? Is there something you want to learn, or again, are you just following instructions? What is the new information or insights that you're going to find useful during your transformation, during your Agile transformation? You know, answer those three questions, then you start getting a sense of curiosity.
[01:42:38] So, wrap this up and quickly now. Um, I talked about engagement models at the start. Um, so this is an engagement model that that I use effectively. I kind of come to think of it as an engagement model. Um, I like to have these five things in place and again, you'll start seeing some overlaps between four disciplines of execution and some of the things I've I've just talked about. Have a true North. So what is your kind of guiding star that's the that kind of sets your direction. What are your aspirations? That's like your wildly important goal. What are your strategies? You're your enabling constraints that are going to help people figure out how to make decisions. What are your tactics? You're you what what is your coherent action? What is it, what is it that you're going to do? And then what evidence are you going to look for that are going to give you some feedback as to whether you're making progress? Those are your lead measures. Um, and I link these to like this. So true North, your your orientation.
[01:43:29] Um, if you are heading towards your true North, you hopefully have some ambition, you should be able to achieve some aspirations, the results you want to achieve. Um, but you're not just going to achieve those overnight, you need to enable those. So those aspirations are enabled by your strategies, your guiding policies. And then you implement those strategies with your tactics, the coherent action and hopefully those tactics, if they're good tactics, they're going to generate evidence of making progress. Um, so indications, um, and then that's going to help you to achieve your aspirations. So these things link together with the true North at the heart of it. And you want all these things to be coherent. There should be some kind of form of correlation between them, you should be able to tell a story about how all these things contribute to each other, but it's not a nice, it's not neat. So Dave Snow talks about messy coherence and I think that's apt here. So Dave Snow talks about messy coherence and I think that's apt here. You don't kind of go, well, this aspiration is implemented by this strategy and this strategy has this tactic and this tactic has this measure and this measure maps to this aspiration.
[01:44:30] It's a kind of a many to many. So you end up with something like this. So this is called the X matrix.
[01:44:37] Um, where you put you put so this this for simplification doesn't have a true North on there. But, you know, our aspiration is to increase our sales, um, to increase our sales by X percent. Um our strategy is to focus on more frequent releases. Um to to implement that strategy, we're going to work on continuous delivery, we're going to implement a continuous delivery pipeline. And we think if we do that, we should be able to lower our lead time. And then the dots in this corners here is basically saying, this there's a strong correlation here. There's a strong correlation here. Um this one down here, it's it's a medium weak correlation.
[01:45:17] So I can start telling a story about how these things fit together. And then, you know, so personally, I kind of think having a single important goal is is you get a little bit too focused. Having a range of of aspirations that balance each other is is my preference. And you're going to have multiple strategies and multiple tactics and multiple pieces of evidence. And we can start visualizing how they all fit together and whether the whole thing is coherent. So one thing you'll notice potentially here is if I look at tactic two, you know what, it it only weakly corresponds correlates to one of our strategies and it's only relates to one of our pieces of evidence.
[01:45:57] So is that the right tactic? Maybe we should be working on something else. Or maybe our strategy is wrong or we're missing a strategy or maybe we we need to find out another bit of evidence. So we can now have a conversation about this about what we're doing, why we're doing it, whether it's working or not, um and ultimately what we're trying to achieve. Nice honest, is it a three? It's on a single piece of paper.
[01:46:21] So, um, there's a template I have a template you can download for this. And again, we'll make the slides available. I think of the X matrix as effectively the thing that really forms coherence. If you have a, if you have an X matrix populated and and this is quite a large one and that I've lots of, you know, five strategies, 10 tactics, I don't recommend filling it. It's just some organizations want to start there and then they learn that there's too much. So and and narrow it down. But having those conversations and understanding your your true North, your aspirations, your strategies, your tactics, your evidence, and being able to to to visualize that, starts generating that coherence. And then we can start forming teams to work on those tactics. So that's when we have uh so the next A3 is backbriefing. So I mentioned backbriefing with Bungay. You give a team a brief, a mission, you kind of go work on this tactic. So this is one of the tactics on the X matrix. They can come back and fill this in and say,
[01:47:18] we understand our context. This is this is the our intent. This is this is the higher intent. The higher intent is actually how does this help the business? So this is your manager's intent. Who's working on this? This is who we need, this is going to work in it. Our boundaries, our freedoms and constraints. So start setting enabling constraints. And then our plan, what are we going to do about this? So this this um is the thing I think that starts bringing some of those constraints in and getting people to thinking around um the the boundaries within which they're going to work and how they're going to work, but we're not down to implementation level detail.
[01:47:54] So then you have a third A3, um the experiment A3 where we're down to the the level of detail of actually having forming specific narrow hypotheses and running experiments. And again, let's make sure we understand to keep your alignment. We understand what the context is, uh, and we understand what the hypothesis and then and the rationale, why we think this is a good thing to do. By having these things on A3s and having the conversations about these things, um we can make sure we we we keep that alignment. So this then is that that cure the running experiments is the way we we we encourage people to be curious. Um, one of the things you'll notice on here is, what results are going to indicate success and failure? And if we're successful, how are we going to follow it up?
[01:48:39] And if we're if we fail, is there anything we need to kind of roll back, backtrack, dampen down? So we're explicitly recognizing failure might happen by mentioning these things on the A3. So that's the curiosity piece.
[01:48:53] And actually, I'm going to throw in a fourth C as a bonus. Um, any photographers in the room?
[01:49:01] No. Okay. Anybody think this is a good photograph? No, it's blurry and horrible, isn't it? I love this photograph. So this is a photograph of my daughter, um when on holiday a long time ago, a good few years ago now. Um, I think this is a great photograph because I was there, so I know what's going on. We were we were on holiday, we were by the pool, they did a big kind of foam party, um, you know, the grownups had a few drinks. The kids were, you know, playing around in the foam. It was a it was a it was a great night. And and because I'd had a few glasses of wine, the the photograph came out a bit blurry. Um, but because I was there, um, I have a lot of context around this. So there's Jeff Paton uses this this metaphor when he in his user story mapping. And I think the same thing applies. Um, and and Tom Thomas El Jackson has this quote from um his book Hoshin Kanri for the Lean Enterprise, which is which is basically the the X matrix book. It's the memory of what was said and felt that creates the alignment, not the final piece of paper. It's like the photograph is meaningful because I was there, I was in in the X matrix, the A3 is meaningful because I was part of the conversation when that was created. So we don't use A3s as a it's just a smaller form of plan that you hand down. It's a collaboration thing. So the fourth bonus C is it's about having conversations with people. All of these things, it's all about having conversations with people. All right, I think I'm out of time. Um, these are the books I've mentioned, um and and referenced, um kind of I'd recommend reading all of them. Um, a lot of what I've talked about, I'm kind of I blog about, so uh those are the kind of URLs for the so strategy deployment is is is really the topic of this. The template you can download for there. And um Ben Lind is very kindly did an interview with me beforehand, um on in Q. So there's a bit more information on there if you want to go and read that. Um, thank you very much, um and thank you to the sponsors.
[01:50:57] Um, I hope you enjoyed the rest of the day. I'm going to be around today and tomorrow morning, I have to dash off tomorrow lunchtime, but um if you want to ask me any questions, just come and grab me. Thank you.

Transcript