 all the ones who have been here during the lunch break, I kind of have to disappoint you. There are all the trumps back there, but it's not me performing any drumming. I guess it's for the evening event, for the reception, and I'm already curious what will happen here. So I'm really glad to see you all here and actual principals. Actually, this is something I have been doing for, so now I do have to do the math, I think for 14 years. I started working staining agile in 2002 and also my first book that I published was on that topic, so that's the agile software development in the large. I know it's just now becoming a really hot topic with SAIF and less and the AD and all of that. To be honest, for most of those now, all kind of frameworks that are offered, I'm a little bit disappointed because for me, scaling agile isn't really anything different than agile. And that's one thing I hope to bring across today to you, that it really relates on how we apply the principles of agile and those help us to scale or do any other difficult things, maybe hardware development or applying agile in a completely different field. So whenever you go back to the principles, you will find some guidance there and I really hope that this is something you will take away from that session. Okay, about the session. So I plan to run this as a workshop and we'll see how this will work. And the agenda is first, yeah, I wanna look at the goals, then we look at the experiences you are all bringing here to this session, then generate insights by examining the principles and then deciding what to do by applying it on difficult issues. So applying the principles on difficult issues and then wrap up and feedback. And for the people who are doing agile for a long time, they might think, oh, I have seen an agenda like this before. Yes, it's actually pretty much what you do in a retrospective. Where you gather data, generate insights, decide what to do and this is it. So you can also do this structure that you know from retrospective for other stuff. So maybe that's another takeaway and have here. Okay, the workshop goal, I mentioned that already. So this is really very, very important to me and I absolutely hope and it's only like every fourth person in the room understands that scaling agile is really the same as just agile, because scaling agile is agile, then I would be happy. So like sub goals that I'm having is understanding the importance of principles for scaling and know how to adjust and adapt to your own needs. So what can you do in order to apply it to the problems you are having? And let's get started for gathering data on your experiences. The way I wanna do this is maybe you know that also which is also used in some of the retrospective, the sailboat metaphor. And the way I wanna use that is that, well for a sailboat, it needs some wind to be pushed forward and there's also something that's holding it back. So the wind that's pushing it forward is the area I wanna first look at and it is what has helped or if you haven't skated agile yet, what is expected to help when scaling agile. So what are the things that you think will be easy to do? And I would like you to discuss that with well like maybe six people and maybe either you arrange yourself around the tables or maybe you can arrange the chairs somehow and just think about with the six people what do you think is helping when scaling agile? What is it that just you think well that has helped me in the past or would help me? And we have three minutes for that and I would like you to take notes and you can take notes already. We have flip charts on the tables so you can take a note there. Maybe you wanna draw a ship first, a sailboat and put the wind there and then, okay, what is it that's pushing that boat? Okay, let's get started. Three minutes, what's helping when scaling agile? The question here is if it's focusing on distributed agile as well. So I didn't have distribution in my mind but if this is the situation you are having here mainly then that's fine too. So my focus right now was more like on just it's big but I know most of the things that are big are often distributed, yeah. If you don't have experience on scaling agile this is why I have what do you expect would help if you're scaling, yeah. Oh, and you don't have to agree with everything. They don't have, you don't have to have consents on what you are doing here. So the one person might say, oh, this is helping and another one says, well, something else is helping. I'm trying to find a way to support this. I'm trying to find a way to support this. Just, I don't know if you are able to see just what are the issues you're explaining and what are some of the things that's coming up. What are some of the issues you're explaining? What's the main thing here? If you now look at what is a really difficult issue but what is the thing that you think is really the hardest when scaling? So it's now looking again at what has hindered or is expected to hindering scaling agile and now what is the main issue? And so it's kind of a sticky issue because you don't really know how to get over that. And it would be great if you could agree on one at your, in your crew. And if you have then two, then it's okay too. But try to agree on one thing. Okay, but I checked with many and I heard that you're done and I would like to see issues you've found. Let's start here, the table here. So we go next then to you. Yeah, communication and collaboration. Yeah, okay, so then, which is kind of everything we are talking about. Okay, so we make here an emphasis. Yeah, okay, yeah. Oh, uh-huh, something else, what about here? Oh, yeah, that's a big one. Two, maybe here. Dependency. So it's similar to this one. Okay, land collocation is what you said, right? Okay, what about here? Distributed team? Dependencies as well. Noticeable outcome. How do I write that? Like this. Which probably, well, which relates as well to the other things. So for example, to the dependencies, which could relate to that. This is why you don't have the outcome. Is there anything else back there? Yeah, which one, yeah. What else? Alignment with external organization. And what was this third one? Training, so which belongs to culture and mindset, right? Okay, can I hear a few of the things, and I'm not sure if I note them down, or maybe I do, of the things that you think that help or that are easy to scale. What is helping when scaling up? Existing knowledge, yeah, some of them at least. What else? Practices and ensuring that they spread. What else? There are like three things then at the same time. So I got this one, organizational support, that's always helpful. And there was something with communication. Pei yori, yeah, right. Yeah, yeah, yeah, yeah. I'm sorry. And I didn't know where to look. Innovation, creativity, helps. A mature portfolio back. So like the big picture, yeah. Oh yes, that's also a big one. Leadership by and probably related to organizational support, alignment, what we had before as something that's difficult. And of course it's helping if you are having it, yeah. Okay, maybe we'll leave it there unless somebody says like, I'm really having something very urgent, that's missing. Motivation, something urgent? That's a yeah, that's a good list. Okay, so experiences, but you are seeing that's helping and also like the difficult things. Well, we didn't go through all difficult things, but through your sticky issues. And we realized already for some that some have a bigger emphasis or some are also related to each other. So they might ask for the same resolution. I wanna go on and first now look at, well, the principles as they are there. So the actual principles as they are defined and hopefully we can use them to generate insights and then later on apply them to the sticky issues. But that's only the third step. So the first thing I wanna look at, well, not really, but you have seen that. I'm pretty sure you have. So the etchup manifesto, don't worry I'm not going over that. With the manifesto, there comes as well the etchup principles. And often people forget to take a look at them. And this is actually what I wanna do here. So the etchup principles, this is a abbreviated list of, oh well, I just did that because I wasn't sure if we have enough handout. So I put them all on one list and it's a lot of slide, I know that. But this is like the full list, the 12 principles as they are defined by the etchup manifesto. You have on the table a few handouts which are just those 12 principles. And what I wanna do now is that you go together in pairs. So we make the group smaller now. So not as big as we had it right now. And you go in pairs and explore the principles by well, taking like two or three of them. Don't look at all of them because it's too much and we won't have time that everyone really looks at all of them. It would be great if at a table you could distribute them so that you're not all looking at the same ones in pairs, right? But it's not very strict about that. And so the question that I wanna ask here and the way how you should look at the principles is how do these two or three principles you are looking at support scaling agile? And of course on the other hand as well, how do these principles challenge scaling agile? So it's a similar question to what we had before which was more just the general, what's your experience? And now we have the etchup principles from the manifesto which define what agile really is. And I want you to look at two or three of them as a pair looking at what's supporting it and what's challenging. And I said there are handouts at the table and I don't know what's best for you. Should I put on the slide the principles back or do you want the questions here? The principles, well, there are different opinions. I put the principle. So what's supporting, what's hindering and in pairs two to three principles and how long, well, at least five minutes I would guess and take notes of your findings. Okay, what I would like you to do next is share your insights at the table or in the crew you were working with before because I hope you looked at different principles. So you can learn more about the other principles, how they help and how they hinder, right? So just exchange what you found out. That's your sharing at the table. You're done, but you're all. At first I would just like to hear back from you if there was anything surprising that you heard when sharing the insights, yeah. It's difficult to put all the principles in the scaling model, yeah. However, if we say, although we are scaling and we wanna be agile, then this is exactly it. Something else that's surprising. Face communication, that's really a big one and it's not easy to solve, yeah. Yeah, some might really be easy to just, you can just do that, right? Integrated software across a lot of teams, yeah. Is anything more like where I think this is really concerning? And do we have one working software? Or do we have as many as we have teams here? And it's not like one coherent thing. That's a, yeah, right? That's where everything surfs around. Oh, the motivated individuals because very often if you look like at the scaled up thing you don't see the individual anymore. It's more like, well, it's a mess of people. Yeah, that's true. It's good that you're asking this now because that leads me to what I think would be helpful. If we now go, well, I'm not sure if we go through the complete list, but it leads to like some of the things that you've found. And now we have the first principle which is the highest priority to satisfy a customer through early and continuous delivery of valuable software. And now the question, one of the questions is already, okay, what is valuable software? Well, the answer from my side is, it's what's valuable, A, from the customer or market point of view. And B, it's also valuable from our business strategic point of view. Sometimes at least that's my trap. Oops, that's my trap where I often fall in. I often look only at the market and customer side, but this is not enough. It also has to be in line with our business strategy. That's also a value. Okay, and if we look now at these first principles, what did you find for the ones who have looked at this that's helping on what's challenging here? Providing the value early and what was it early and continuous delivery? Yeah, okay. And if I'm now playing the devil's advocate, how is that different from in the small? You probably have the same issue there. Maybe not, but this is the way you're just saying what is the challenge here sounds to me like, okay, that's a challenge we are having independent if we are scaling up or not. Is that true? Okay, thing else around that principle, the first one. Yeah, exactly, exactly. So it really, there are at least two things that go with that. One is finding out what will really provide a value to the customer and to our business strategy. So back to your question, it's fine, we'll get you. That's one thing that might be difficult as well in the small, but it's often harder, the bigger you are, because then also the more is implied by the product you're building. It's just harder to understand where is that, I think this is what you had here. It's a big picture, it's harder to understand. And on the other hand, it is also what you said that the point that are we really able to deliver early and continuously because if we have a lot of people, a lot of teams there, how does integration really work? Does the build really always pass? How long is the build time? And how long do the tests take in order to verify that we really have a valuable software that's working? So integration is a really big topic here, which needs to be solved somehow. If we go on in order to really understand what the principles are asking from us when we scale up, what about the next one? Has anyone looked at the next one? Exactly, maybe I read this. So welcome changing requirements, even late in development, HR process is hard to change for the customer's competitive advantage. So you're saying the biggest point here is that one, what you found out about the sticky issue already. And now the principle says, okay, but if we are agile, this is what we do, right? So we still wanna welcome the changing requirements. Maybe what you need to do upfront is providing a structure with the team that allows to make those changes. So looking at the dependencies from a different angle. Very good, yeah, that sometimes is the case, right? Very good. Very good, excellence. That's a very great contribution here because it's very often so, if you look at one principle only, it might shift you into a direction. If you overdo that, then it's not doing really any good. So for example, if you really kind of welcome the changes all the time and you are not taking care of the quality of the system, then you create just a mess, right? So it doesn't, so very often one principle can't go without another one and the two you picked are really great ones. So the attention to simplicity and to technical excellence, right? Is there anything else around the changing requirements that you came up with? Should we move on? Yeah, there's one more. So that the reworks, which is, again, the quality in which way we deliver it at firsthand, right? Well, of course it might as well be, we have understood something in the wrong way, but very often at least the way you're saying that it's maybe we deliver something that's buggy and that's why it's safe in the room right now. I once learned for any kind of session disturbance is take precedence. So if I now just do like nothing has happened, it will not really work. But I see this one is coming up and here something is coming up. Let's see if I can make the slides. Something is happening here, but not there. So I see it on the monitor, but I don't see it here. It's coming. So I have to be patient. Is this what you're saying? I'm not always good at that. Okay, so yeah, it's coming and so I trust it will be working. So we set the rework and the buggy delivery might be one thing that's changing, welcoming the change. Uh-huh, yeah. Well, it's actually independent if we are in the small or in the large, but it's a very important point than the point is in case you didn't understand. If you're welcoming changes all the time and you're not saying like, okay, this also has an impact on what we are doing, then you might not be able to hit the deadline, right? Or you might not be able to deliver what all has been committed to because there's always those, there are those changes coming in. So the thing is probably the next, is it the next? No, the two after the business people and developers must work together daily throughout the project is very close to related to that because it requires an understanding what that change really means because we are busy implementing that change. And if it's really so important, then this is fine, but it might have the cost that something else will not get implemented at least not in the same amount of time, right? Okay, let's go to the third one, which is maybe, maybe what I should do is going back here now that we have the slides and so you see this better, right? If we go to the third one, which says deliver working software frequently from a couple of weeks to a couple of months to the preference to the shorter time scale. Anything you found here that makes scaling easy or challenging, the scale team is supporting this? So what I'm, so in general, focusing on that principle you're saying is supporting scaling up? Yeah, I absolutely agree, I absolutely agree. Anything else to that? Actually, I think I take a different approach here, which is I think you don't have another chance really. And I know there were times and still there are some people discussing that and saying, well, but if you are in the large or if you are distributed or whatever you are struggling with, then this takes more coordination kind of also the collaboration part and therefore, that's what people are discussing sometimes, therefore we wanna have longer iterations or sprints for reducing that coordination effort, but of course I guess, I guess of course for you as well. The longer the period is that we are not coordinating the harder it will become, right? To really collaborate. That's also why I see this from the other angle and saying, okay, the higher the risk is that we are having the shorter the iteration should be because with every iteration or every sprint, if you will, we get the feedback and only with the feedback we can A, learn and B, correct what we are doing. And of course we wanna know also or get the big picture on a daily basis that at least to my experience, often have to do this really every day but at least on the iteration length to get the feedback over what's happening with all those teams, with all the development effort in that project product development you are in. So it's really, as you said, it's supporting, scaling in a way that it's reducing the risk. And if you will, everything we are doing in actual is about getting feedback. And this is kind of the biggest thing with that. So the feedback on the development effort, of course we have also other things like the feedback about the valuable software that the customer which was the first principle for example. That's it, that's it. Answer it. Well, I said I take a different angle to that. So I think there is no question. That's true. However, if you can integrate and you can deliver and you can test it, this means, I'm sorry, but you're not doing really anything because there's nothing, who was that before? Someone said this with the working software as the primary measure of focus when I just did like a first round of surprising and whatever. If there is no working software then there isn't anything, right? So, and that's why I said I don't see any chance around that. So I think this is what we have to focus on and speaking of automation, yeah. It's a big topic. That means sometimes that we have to put all our forces on automation and not so much on implementation of new features, although of course sales and marketing always wants us to focus on implementing new features. But if we are not good in integrating, delivering, testing and all of that, then it doesn't help. We can implement as many features as we want because we will not be able to ship them. So this is actually the thing that helps us staying in the market or getting a food in the market. Yeah, it's a big thing around automation, that's true. Anything else about that or should we move on to the next? So the next one is this with people and developers must work together lately throughout the project. Did anyone have that principle and what were your findings? So they can be far away or they are too busy. Actually, this is also something I sometimes see in the small as well. So it's not always only a scaling up problem, but it's harder when scaling up, absolutely. Same thing, yeah. And still the thing is if we don't do that and we do all the other things we have been talking about, then we might build stuff nobody really wants and we have a similar problem than if we don't ship at all. Well, what you need to do is really to ensure that every team has access to business people and there are different ways how you can arrange that or manage that. So one thing is, for example, if they are far away then you'll have to train people at your site who can answer that and who are in closer contact than on a regular basis with real business people. So more a shadowing concept also. But ignoring it and saying, well, okay, we don't have the people that doesn't help. There is the, you need to put the motivation. Did we have that somewhere? It was here, right? Motivation helps. Build projects around motivated individuals, give them the environment and support they need and trust them to get the job done. What about this? Did anyone take that principle? If not, then we can also skip it and go to, from my experience here, I think this is very much also related to self-organization, the motivated individuals and it is of course related to leadership and management in general. And yeah, mm-hmm, mm-hmm. To me, the key thing here is, maybe it's too obvious for you, I don't know, is cross-functional teams. Because only in a cross-functional team, you are able, as a team, A, to take full responsibility about what you are building and B, see the result. And this is something that I think is the thing that provides the highest motivation to people. If they see what they are really building and it's not like I'm building here the UI but some team in a different country is doing something with that UI and making some value for a customer out of that and maybe there's a third team implied with that. So for me, really the structure of the team is the most important thing here. And again, it goes together with the self-organization. So having cross-functional teams who can take over full responsibility for a feature, really. So if it's smaller than it's a story and they are not dependent on somebody else for finishing that story or that feature, this provides the highest motivation. And also it's the key to self-organization. So that's my take on that. So the structuring is the main answer here. An interesting point, so the question is as a newcomer in a scrum team, it's harder because it's cross-functional. And actually, this is not my experience. That's the thing that I hear the most often with my clients that they are saying they are really surprised when new people are joining how quickly they come up to speed. And it is because, again, the motivation. I assume, because the team as a team holds together and is interested in also helping the other person to understand. And there's almost everyone who can show something and mentor. We don't have the time to self-learning. Well, that's the whole question about how do people deal with velocity? Which is actually not really part of, well, it is part of the sustainable development, the number eight here. But it's not really in the principles, but often people understand in a team velocity as productivity and not as a capability planning tool. And if you look at this as a capability planning tool, then you plan learning in. So that's, I think, kind of a wrong understanding. However, it's not related to scaling. That you have that as well somewhere else. And again, the thing that I'm hearing of and what I'm experiencing is that in cross-functional teams and in scrum teams in general, it's much easier for new people to come in and to learn what's going on and be helpful almost from day one in some way or the other. So that was the motivation part. If we, let's check the watch. And if we move on to number six, the most efficient and effective method of conveying information and to end within a development team is face-to-face conversation. That was the big one that was mentioned right away. That's, wow, that's the challenge if we are scaling up. Okay, so what does it mean? Of course, scaling up. Frequent communication, absolutely. So it's in the principle and maybe you say afterwards when you leave here and say, oh, Utah takes the principle as a Bible or something. And on the one hand, maybe I really do that because I think either we wanna be agile or we don't wanna be agile. And so the principles say this is the kind of the definition of agile. And therefore, if face-to-face is really the thing that's so important, we have to ensure that something is happening here. And as you're saying, for example, periodic conversation or communication is one thing. And did you mention that earlier? Somebody mentioned it here. Was it you as well? That this is the thing that's helping? Yeah, absolutely, I agree. I agree completely. It's, well, actually for probably almost all the principles, I'm not saying it's easy to apply them. However, what I think if this is something that's important for agile, then we have to strive in order to make it happen. And one way, for example, what you're saying right now is more like a distributed setting is, well, there are different answers to that. But one answer is there are times where we meet as well for a distributed setting. That's at least in the projects I'm working on. And we ensure that whenever we meet, which might be not frequent enough, but there are times we meet at different locations. We are not meeting at the same location all the time. And just because of that, different people have a chance to get to know to other people on the team. Because not everyone will be traveling, of course, if it's really large. That's one thing. The other thing is, for example, the closer we are together and we are large, the more frequent we meet face to face, but the shorter the time frame as we are meeting. And the further apart we are, the less frequent we see each other. But the longer the time frame is, the duration we are seeing each other. It can go up to ex-patriots, yeah, who work at another site for a really long period of time in order to understand the different mind-side culture ensuring the connection between sites. Or it can just be that, well, some people go to another site maybe for two weeks. So something in between there. And speaking of, like now I'm from Germany. I'm now in India. If I have a team in India, which I actually have right now, then if I go there, I don't go there for one meeting, one hour, or probably not even for a day. The least I'm going there is for a week. And if it's something that's quite close, so maybe two sites in India, then you can see each other more often because it's not so far, but not for such a long time, of course, not for one week. So that's kind of the answer. It's still important to ensure face-to-face. And still at least what we are doing is we strive for making it happen as good as we can in the framework we are in. And it's always on the top of my list of the top-high risks that if we are not focusing on it, then probably other stuff is not working. Well, no wonder that this was mentioned first, communication and collaboration, because if this is not working, probably integration and automation will as well not work because everyone is doing something different and we are not really working together. That's why it's so important. Working software is a primary measure of progress. I know we have talked about this already, so I just wanna check, is there anything that we haven't talked about yet that you, if you have discussed that at all? So who focuses on integration and testing perspective? Of course there are, with a lot of things, there are different answers to that. So I made good experiences with, A, we have someone per team ensuring that, and it could be that it's not a person, but a role which is changing amongst the team members. So we have in every team, at least one person ensuring, for example, for a sprint, that integration is working. And therefore, because we have somebody like that in every team, they get together to be able to leave if there is any problem. And also work on automation. That's one thing. Another thing that we have done successfully was having a separate team, focusing on integration and build only. So on all the automation around that. So it's not that if something isn't working with integration that they have to fix it, that's not the thing, but ensuring that integration and build is automated and it's fast. So it's optimizing integration and build. So these are kind of the two strategies that we used. And actually a good idea might be if you're struggling with this a lot, so maybe your first step is to come up with a team doing that, so that you have a good foundation for it. And then dissolve the team and decentralize it to all the different scrum teams. So sometimes you have like different, depending where you are right now, different steps in between, which relates of course to working software. Okay, did anyone pick the HR processes, promote sustainable development? The sponsors, developers and users should be able to maintain a constant pace definitely. I would think so too. And I don't see, I don't see actually something that's specifically different than in the small. At least right now nothing comes to my mind. So it's, if you're struggling with it in the small, you might struggle as well in the large, but you can take it as a leverage for scaling up. We talked a little bit about the other two, the next one is continuous attention to technical excellence and good design enhances agility. Any insights about that? Yeah, agree. The thing that I see sometimes is that people don't take, don't pay attention to that. And they struggle with it on one side with one team already and then they scale up and they wonder, right it's not getting any better. So that's something you might struggle with anyway and the problem just gets bigger and bigger the more you scale up. So exactly, absolutely true. And the major thing and you probably have heard about that too is the communities of practices that help as well spreading the knowledge about how to do this and that. So how to focus on technical excellence or what kind of tests and so on to use for that. Simplicity, the art of maximizing the amount of work not done is essential. That's true, you plan for it and I would like even to add to this principle that simplicity is essential for scaling up. And the reason is again the thing that you've brought up here with the dependencies. If you don't focus on that then this will get harder. And on the other hand what I've seen as well is if you are not focusing on simplicity it's also harder to understand what's going on in the system for everyone and the bigger the effort is so the more teams and people you are having here the more likely it is that there are also some people who might not understand it. So complexity is an enemy for scaling up. It's really like here, cowboy coders also they really hinder scaling up. Okay, I think we talked also about the next one at least a bit, the best architectures, requirements and design emerge from self-organizing teams. I talked about self-organizing teams before however not about the other part of the architectures, requirements, designs. Anything about that, did anyone pick on that? Talk about that, probably most talked about the upper ones. It will not work, I agree. Really say yes to that and the reason why I'm not saying yes is because I think it depends very much on the team structure and the thing that says here the architecture, requirements, designs where you see this is kind of everything that's involved in creating software speaks already about cross-functional teams and that's the key thing here really. If you do have cross-functional teams all over the place when you are scaling up then it's not really any different from a team perspective. What you need as well is then the communication across teams but still from the team perspective and ensuring architecture, requirements, designs and then if you also have communities of practice that's not a big problem. Absolutely. Let me just finish the next one and I wanna get back to this because then we are through with this. So the last one is at regular intervals the team reflects on how to become more effective than tunes and just its behavior or calling lead. Anything about that one? Any specific retrospectives? Still there are important and the learning happens with the retrospectives and of course the thing might be if you are completely distributed that you need some tools for doing that or if you are actually this session wasn't really focusing on distribution more on scaling up but still if you have a lot of teams then you need to think about as well having retrospective at the scale up level. And running out of time. I still have this in mind what you said about aligning the architecture and I still think what we do now that we have talked about the principles to look at deciding what to do and what I would like to do is to look at your sticky issue at your table and we take five minutes to look at how do those principles help solving that one sticky issue that you came up with. So and you can pick just like a few a couple of principles and find out how they help you there. And I wanna spend really only maybe three minutes on that and then we wrap this up and also I wanna talk about the architecture problem again so if you look at your sticky issue and the principles how we talked about them now and see how they help for your sticky issue here. Three minutes to that. My question is did you find now after we talked about the principles something that's helping with your sticky issue some of you are still in discussion. So did you find that any of the principles help resolving the sticky issue that you came up with? The dependency was yours. So if you focus on the technical excellence it helps with dependencies, right? That's what we talked about beforehand. And yeah, okay, yeah, you were first and then you. So the motivation promise your answer to that. Actually I would think the answer to that is as well focusing on really the actual values and understanding them. What about integration helps to resolve the dependency and then again we are at your favorite topic automation. Anything else that you found? Principles that help with your sticky issue? So you are addressing it with collaboration and structure of the teams, I understand that. Right, anything from here or was that mentioned as well already? Okay, so there was still this open question about the, well question it was more a comment on the architecture and how to create the alignment. Well we have here external facing alignment. So for the architecture, actually the answer is not so much different than what I gave already for integration. So first of all the answer was well it depends what is your biggest problem here right now and also the situation you are in. So for integration I say okay if you are not really able to deliver then maybe you need a team that focuses on that and ensuring that the build works and is fast and is automated. And so the same answer is here too. If you are really struggling with the architecture you might need a team that's providing a service which is creating the architecture for the other teams working on that architecture. And once that architecture is in place and maybe the complexity is not really high in your system then you don't need this but you have more a decentralized as I said with the integration and you have like people in every team or it's a role in every team that ensures the adherence to the architecture rules so that you have conceptual integrity here. So it depends on how vague is the architecture, how much, how complex it is and how many changes you are having here. And this depends on the support that you have to provide and the kind of the lower spectrum you don't have a lot of changes in the architecture your system isn't really complex then it might mean that you only have a community of practice coming together at times and sorting things out. And if you are at the other hand of the spectrum and you have a lot of changes going on there in the architecture then you need really people focusing on that only because if you don't this would hinder everyone else to work on their stuff. So the answer is for a lot of things, it depends and I know we are almost done here and I would like you to take maybe 30 seconds and reflect, what is this? Yeah. So to conclude about the principles it is for scaling up the principles are still valid and that's why I made that entering statement that scaling agile is still agile. It's not really different. However, applying those principles might be harder and actually what we didn't really talk about, well we talked about it indirectly often we think they can't be applied because we are scaling up because some other rules have to put in place which is not true. They are still, these are still the guiding principles for agile no matter if we are scaling or not. And just the example with the face to face it is hard to do that and you will not be able to do this on a daily basis but if you understand that it's important in order to be agile then you will ensure to have it as often as possible and in different ways as I said like just different meeting locations and ensuring the face to face this way. So applying the principles still hold true and actually now the time is up anyway and I wanted to do maybe this is your little homework to reflect on what is your key takeaway from that and maybe at least I heard this kind of here. My goal was not to talk about like less unsafe but my goal was to understand what is the kind of the definition of agile and how does it still hold true in scaling up and that there isn't really anything else we need. So thank you very much. I know that we are to the end here today and I hope you have some takeaways from that session from that session here and enjoy the rest of the conference. Thank you.