This is a transcript of episode 159 of the Troubleshooting Agile podcast with Jeffrey Fredrick and Douglas Squirrel.
Drawing lessons from Squirrel’s driving test experiences, we explore why making a cultural or process change that’s intentionally suboptimal can be surprisingly valuable.
Squirrel: Welcome back to Troubleshooting Agile. Hi there, Jeffrey.
Jeffrey: Hi Squirrel. A couple weeks ago, we talked about driving lessons and today I understand you can tell us about your driving test, which sounds great.
Squirrel: I am indeed. So this actually has relevance to the topic at hand, namely troubleshooting Agile because I was coaching a client and normally I have to anonymize the clients quite a lot. This one I don’t have to anonymize because their problem is universal. And the problem they were having is that they were preparing hugely for each thing. And the person I’m coaching was saying that he was ‘all ready to do something new.’ And I said “Great!” And he said “So we’re holding the first meeting tomorrow and then we’re hiring the team and then we’re doing the next.” This was going to be three months from now before anything happened.
Squirrel: I’m going to get back to the driving test. Don’t worry, I’ll explain the driving test piece as it comes up in this coaching interaction, because I said, “so why don’t we do something this week? So your homework before we get back together is to take action on this.” And he said, “but wait, that’ll be wrong. We’ll do the wrong thing.” And I said, “let me tell you about my driving test, because I had a very friendly driving instructor who was helpful to me in various ways. And one of the ways he was helpful is that when we would go for the driving test, news flash, it took me multiple tests to pass. I’m not a great driver. And he said, “now what we’re going to do before we go to the test location is we’re going to drive around a little bit. You’re going to drive around a little bit so you can make all the mistakes first and you can get them out of your system.” So I told this story to the person I was coaching and he said “that doesn’t help me at all. Why do you want me to do this thing that’s not going to be right?” And I said, “well, this is part of driving. This is part of moving quickly.” And it’s exactly an Agile principle that we already at least profess to do in our Agile teams, namely, let’s try doing something that we know is not going to be right, but from which we will learn a lot. And he said, “OK, so for this initiative-“ this was an internal initiative shifting their culture, doing something in a new way, “we could do the same sort of thing, like an AB test or an experiment that we would learn from.” I said, “exactly, you’ve got the right idea.” So just like my driving instructor said, “well, hey, let’s drive around a bit so that you get out all the bad ideas.” You remind yourself what all the things are that don’t work, then your Agile team can do the same. Try something without all the planning, without all the preparation. And, you know, it’s going to be wrong. You’re going to get the wrong result. And that’s a good thing because it’ll give people something to argue about. It’ll give people something to argue against. It’ll give you examples of actual behaviour. Now, there are some caveats to that, but I want to see what you think of it first Jeffrey, was my advice good?
Jeffrey: Well, I love it, it’s fantastic! Which is why we’re doing the podcast on it, of course.
Iterative Learning and 3D Printed Fish
Jeffrey: What I love about this is it gets very much to one of the dilemmas and one of the powers of Agile. The big element in Agile for me has always been the idea of iteration. And as you say, we know we’re not going to be perfect. We know what we’re doing is not right. But we also we’re going to learn from what we do in this idea of pain to learn. And being OK with being imperfect is so important. And it brings to mind the difference between iterative development and incremental development. And there are very, very important I understand the difference. Incremental development is something that people often aim for when instead of iterative. And the difficulty is incremental as I have this idea up front and I’m going to deliver everything according to my vision in pieces. So I’m doing incremental delivery and that’s better than Big Bang delivery. But there has the weakness that it’s it’s not still optimal. It’s not optimised for learning it.
Squirrel: Let me just check I understand it right, so this is kind of like a 3D printer that comes along and it has a picture of what the I don’t know, the fish or whatever it is that you’re going to make on your 3D printer is going to look like that first it lays down the bottom of the fish and then the next, scales and then the fin and then the next part. And so it’s always adding more to fish. It’s never going to turn into a whale or, I don’t know, a dish or something. It’s always going to turn into the same thing coming out of your 3D printer because it has an ultimate plan and it’s building that. That’s incremental, if I got it right.
Jeffrey: Yeah, exactly. That’s perfect. And the other view is iterative, where I’m going to go ahead and rework essentially the whole thing. I’m delivering something in low fidelity and then I’m going to go ahead and essentially do it again. Potentially I’m going to deliver something that may be different than what I was anticipating. And the idea here is that I’m more open to learning, I’m not as fixed. On where I’m going to end up, so it might be, I don’t know exactly how to do it- in the world of an alternative to three printers, but I might be thinking like I’m trying to make a shape to go in the water and I’m going to be refining it, refining as I go until I end up with something nicely fusiform and maybe end up with a dolphin shaped.
Squirrel: Well, I might imagine the Wright brothers inventing heavier than air flight at Kitty Hawk. They started out with a bunch of parts and they put them together and the plane didn’t take off and then they put them together a different way and the plane didn’t take off and they kept going until eventually the plane kind of lurched into the air. They didn’t come out with the idea ‘Oh, yes, we’ll curve the wing just this way and we’ll increase the curvature by this much every day. And here’s our detailed plan for how we’re going to launch on the sands of Kitty Hawk’. They said, ‘hey, we’re going to show up with a bunch of parts and we’ll keep tweaking it until it works. That’s the sort of thing that I think we mean by iterative development or in my case, iterative culture change.
Jeffrey: That’s right. There are iterative ideas I think that people have heard about in the context of software development. However, what you’re talking about now is taking it, making it more generalisable to say, you know, “actually we can do the same type of, you know, non big bang approach for things other than software. And I think that’s the part that I find really intriguing here.
Pick a Direction and Start Walking
Squirrel: Well, great. Well, I hope listeners do, too. So let me give you some caveats and some guides to how to do this. So the first thing to do is exactly what I assigned my coaching client to do, which is come up with a bad idea, one that you know is wrong. Now, what you’re not sure about I mean, if you absolutely know exactly the right thing to do, then why are we even talking about it? Right. So just go do that.
Squirrel: So we’re assuming that you’re in sort of in Cynefin in terms of the complex domain, you’re in a situation where you’re not sure what the right direction is. If you’re interested in Cynefin see the show notes where we’ll link to to the Welsh word Cynefin, which is spelt funny. But assuming you’re in that situation where you don’t know quite what the right thing to do is, do something that you know is wrong but is in the right direction. So you wouldn’t want to make a cultural change, I don’t know, toward authoritarianism and disempowerment if you’re trying to get your team to be more creative in their software development, that would clearly be the wrong direction. But you might not know how they could be more creative. So you might say, “OK, we’re going to release something this week that we didn’t think of at the beginning of the week. That’s going to be our way to be more creative”. We’re almost certainly going to release the wrong thing and this almost certainly isn’t going to be the way that eventually will be more creative. But what we’re going to do is do something that we know is wrong but is in the right direction so that we can see which bits of it are wrong and which bits of it actually turn out to be more in line with where we’re trying to go.
Jeffrey: You said “that we know is in the right direction”. I want to challenge that a bit because I don’t know if it’s important that it’s in the right direction because I can imagine using this approach that you’re using where you don’t know what the right direction is.
Squirrel: Sure. Sure. Absolutely.
Jeffrey: And I think as you’re doing an experiment where you can learn, “oh, my God, this is totally the wrong direction. We should we should try going somewhere else.” Obviously, as you say, we’re not saying don’t use any of the knowledge you have.
Squirrel: Exactly. That’s all I mean is do something using the knowledge you have that sort of seems like it might possibly be right. But you might intentionally do it in a way that you are pretty sure does not dot all I’s and cross all T’s.
Jeffrey: Yeah, I think that’s one thing we can be certain here very often is we know that what we have won’t be totally correct. It won’t be, you know, perfectly the right thing. And that’s what we’re arguing. What I like about this is to unblock people, to start making progress sooner when often the tension, the draw is, well, wait let’s go slowly and try to analyse this and try to, you know, let’s make sure we don’t make a mistake. Instead we’re saying, “no, no, make a mistake. Make a mistake as soon as possible because you’re going to learn from that mistake.”
Make it Cheap to Fail
Squirrel: Exactly like my driving instructor, he said, “let’s go out and get the mistakes out of the way, lets eliminate them early.” Now, there’s a caveat to this and it’s one that can help you feel more calm, because I know some of our listeners listen to us when they’re driving or when they’re jogging or something. So try to stay on the road or the pavement, you know, wherever you are and some of you are wigging out right now, “oh, my God, you’re going to tell me to do the wrong thing and my boss is going to kill me and so on.” So the other caveat is to make sure that whatever you’re doing is reversible so you’re not doing an irreversible change. So if, for example, I wouldn’t suggest that you fire half the team because you think it might possibly be that they have the wrong cultural attributes and you’d like to replace them, that would be an irreversible step and probably would not help you to recover if you suddenly find out. Actually, most of those folks are people that you want in your team. So that would be the opposite of what we’re suggesting. That would be a bad idea. That would not be powerful, well, it would be pretty powerful, but it would be in the wrong direction. So what would be much better would be.
Squirrel: So let’s suppose that you think that your team could be, I’ll stick with that more creative example, you think your team could be more creative, but there’s some people who are less creative then what you could do that’s reversible is form a sub team for a short time and say, “OK, for this sprint, we’re going to split into two and we’re going to have kind of the business as usual and we’re going to have the more creative and we’re just going to see if that works”. And that might have the effect of annoying some of the people in the business as usual group, confusing some customers, slowing down productivity. It might have various negative effects, which you could learn from, might also have some very positive effects like you come up with a new idea you never thought of before, and you learn that you can encourage creativity in these ways. But then after that sprint, you can always go back, it’s a reversible process. And so I’d suggest that if you’re going to apply bad ideas that you use the knowledge that you have and that you make sure that the steps you take are reversible.
Jeffrey: This point about reversibility reminds me of the phrase that people will often say, which is, “make it cheap to fail.” And we’re talking about this reversible versus irreversible that’s what comes to mind for me. That example you use of firing half the team in a sense it is reversible, you could go back and rehire, if not exactly them, their equivalents but that would be very expensive.
Squirrel: Yes, very good point.
Jeffrey: And so structuring your experiments to be cheap to fail is a good idea. The other thing that struck out to me is this idea that we’re embracing the idea that we’re doing something that we know is going to be wrong, that we know is going to have mistakes in it. And what I like about that is because the question is, what’s the alternative? The idea is, we don’t really have alternatives to prevent mistakes or rather not all the mistakes.
Squirrel: And going slower, in my experience, usually doesn’t prevent mistakes. It actually tends to change the dynamics, change the situation. The world changes in the months that you take to do something different so that your precautions and your mitigations don’t work and you miss all kinds of opportunities to learn in the interim.
Jeffrey: Yes, I think that’s it exactly, the lost opportunities to learn. And because I think when we do this kind of planning, we’ve got everything together, I want to say, it’s not that it’s useless. It’s not that we don’t prevent a certain class of mistakes. But I think it clearly, in my experience, doesn’t catch all the mistakes, it doesn’t catch all the problems. So we’re going to have this period of learning through doing anyway at some point we’re going to do something and we’re going to learn at that point that there were things that we missed. And the question is, when do we embrace that process? We can start tomorrow, we can start today or we can we can do it in the future when we’ve kind of gotten rid of all this low hanging fruit, all the ones that we were able to think through. And then we’re left with the ones that we couldn’t. To me the lost time on things that we could have learnt very quickly is where the problem comes in.
Squirrel: And then, of course, one of the values, one of the things that you can get as a side effect, one of the valuable results of this is I’d certainly advocate that you share exactly what you’re doing. This is what I told my client to do. I said, “now tell people that you’re doing this and you know what’s wrong and make sure that they’re aware that you’re intentionally doing something that is less than optimal, that you haven’t done all the planning.” And that’s a very useful cultural message in itself, because that can lead to them taking that sort of risk, again, reversibly in the direction of the knowledge that they have in their own activities and in their own organisations below you. And that can be extremely helpful as well. So you’re giving a cultural message, I’m OK with being wrong, it’s OK if you’re wrong. I would rather learn quickly as an organisation than be perfectly right. And taking some risks of being wrong is a good thing that we do here in this company. And that, I think, can be a very useful message to pass along as well.
Jeffrey: I really like that message. You might tell I’m pretty excited about this whole idea and I’m surprised I hadn’t heard the story from you before. Amazing. We’ve been doing this for a couple of years now and there’s still new stories coming out. It’s not like you just took your driving test in the last couple of years.
Squirrel: No, I’ve been driving for a while. I’m not sure I’ve got a whole lot better. I still have a lot of mistakes to make, but I at least managed to pass the test after multiple attempts, let’s put it that way. I had I had many chances to try again and that was OK. I learnt from each one.
Today is Not Too Soon
Jeffrey: One thing that resonates with me that’s been going on recently at work for me is we’ve been talking about kaizen, continuous improvement and Toyota Kata has been back in our vocabulary. We’ve been doing this deliberately. And one of things I always liked about Toyota Kata, was it has the question saying, go through you go through the process about identifying what obstacle- for people who don’t know, it’s a small enough process describe it in a bit more detail. You have some objective that you’re trying to reach, some end state that you have in mind. And then you have your current condition, where are we now? And then you choose an intermediate target condition. So this is something that you believe is on the path towards your ultimate vision or the direction that you’re going. And then what you do is you experiment through PDCA cycles. They call them plan, do, check, act, from your current condition towards the target condition. And the thing about it that I like, one of the elements I like is you choose one obstacle to work on and then you say, ‘OK, so what are we going to try?’ And this is part of it, these are experiments that may not work. ‘What are we going to try? What do we think is going to happen? And then there’s the question of when can we see the result?’.
Jeffrey: And in the book to Toyota Kata, Mike Rother has the line, he says, “and today is not too soon.” I love the idea that you can have this sort of discussion and be like, what time is it now for us? It’s a quarter to four. When can we see the result? Would it be possible-.
Squirrel: How about by five?
Jeffrey: Exactly. Yeah, could we know by five? That would be OK. We don’t need to wait. It doesn’t need to be a long process. We can start learning right away. And I really like that about the Toyota Kata mindset and I really like it about this message that you’re giving, which is, go ahead and make the mistakes and also make it OK for everyone else. If we want to be a learning organisation that means it has to be OK for us to make mistakes because learning is the detection and correction of error. And so it’s only by when we try to do something and can detect that gap between what we expected and what reality is that we have the opportunity to learn. So, yes, I’m very excited about all this. And I’m very happy that you brought that in as our as our topic for today.
Squirrel: Fantastic. Well, I hope listeners also find it useful. Please go forth and make some mistakes. We’d love to hear about it if you do that and make them today, don’t don’t wait. So if you’re going forth and doing that, you can get in touch with us and ask us questions about it. Tell us about it. Tell us that we’re wrong. All of those things are very welcome. You’ll find all the information about us, Twitter, email, carrier pigeon, anything else you want at conversationaltransformation.com. Things like troubleshootingagile.com will also get you to us and we’d love to hear from you there. There’s also free videos and messages about events. We have a very busy March, all kinds of stuff happening then. So sign up for those. Find out more about us and get in touch with us. Most importantly, if you’re out there accelerating your learning by trying bad ideas. Jeffrey, I’ll see you next Wednesday. All right. See you then.