Organization Design and Product Leadership: Real Change Under Pressure

Ather Nawaz, Aitor Gomez Matias at UXDX EMEA. Video: https://youtu.be/6_SnT1m_3Bc

Readable transcript: edited from the recording's captions for readability (fillers and false starts removed, punctuation and section headings added). Wording is the speaker's own. Timestamps are positions in the video. Names marked [?] could not be verified against the audio.

Your org chart has a user experience

[00:00:08] Aitor: Hello everybody. What you can see on the screen is our vision, the vision of our company: let's create a world that runs entirely on green energy. And I'm very happy that, after all the talks we've had today, all the keynotes, the vision stands whether or not we work with AI. Today we are going to talk about how organizations can affect the product experience of your users. The product experience is built before the users get the product and get to use it. That's what we are going to talk about: how organization design affects that. My name is Aitor Gomez. I work as a transformation lead, and I'm very happy to share the stage with Ather, product management coach.

[00:01:01] What I want to share with you is that when we design organizations, we might be designing for handoffs, we might be designing for slow decision-making. We might be designing for many things that can affect the experience of our users. And the question is whether we are conscious about the way we design our organizations. Ather, could you tell us a little bit about what we do at Ørsted?

[00:01:34] Ather: Thanks, Aitor. Hello, welcome. Just to give a bit of context of where we are coming from: Ørsted is a global leader in offshore wind, and it has a big history, around 10.2 gigawatts of energy that we have produced or are producing as of now. If you want to put it into numbers, it's close to 8 to 10 million houses that we are providing energy to. So that's the sheer scale of the energy that we are producing using offshore wind.

[00:02:17] We also built the world's first offshore wind farm, back in 1991, so there's more than 35 years in this area. And a couple of years back we built the world's largest wind farm, in UK West [?]. So when you look at the sheer size of the company and the production that we are doing, you need to understand there are many decisions that do not happen in one group. Some of those decisions touch several areas at once. That is the canvas that we are trying to design on, and there could be many problems in this whole process.

[00:03:09] Aitor: Yes. Before we get started and go deeper into what we have done, the learning, the journey that we have taken in Ørsted IT to improve the organization and the structures, which led us to form a new organization called Digital Technology that is now embedded in the business, I would like to ask you the following question. How many of you feel, or have felt, that the organization where you are working right now or were working before is getting in the way of your work? Can you please raise your hand? I know it's difficult after lunch. Okay, a fair number, perhaps 50%.

[00:03:53] So we are in the same space, or we were in the same space, and I'm very happy to see that the talk might make sense to all of you and that we are not the only ones. The saddest part of it is that we are not the only ones who feel that the organization is getting in the way of our product work. Our users will also feel that. The way I like to think about organization design is that your organizational chart has a user experience embedded into it. The users can feel in the products that they get whether there are many waiting times, whether the collaboration across the teams is not flowing. They also feel if ownership is unclear, if priorities are unclear.

[00:04:52] We call it many things nowadays. I have been working in this area for some years already, and we call it many ways. We call it dependencies, we call it complexity. The typical thing we say is that our product is more complex than the product of the competitors, but the truth is that our users do not care. At Ørsted we build many digital products, and those products are being used by our users, and what they want is to operate, or to perform trading, or to do some work in finance. These people are doing very important work and they are very capable, but sometimes their experience is fragmented.

[00:05:41] And it is not fragmented because they cannot use the products. The experience is fragmented because the organization design is fragmented. The way that we group the teams, the way that we fund the work, the way that we take decisions will affect the product experience of our users.

Pressure has a cost and a gift

[00:06:01] Before we move on, I want to be very honest with you. This is not a blueprint that you can go back and implement. What we would like to do is give you some principles, some ideas, so that when you go back to work on Monday, you can perhaps ask different questions, or talk with your colleagues or the responsible people about organization design, about what could be done better. We are in the middle of the journey. We are still learning, we are still delivering, and we are still designing the organization. The word learning is very, very important. We need to include those learnings and listen to our users, but also listen to our employees, in order to design the organization further.

[00:06:48] So this is a story about how we are designing the organization more intentionally around product leadership, around value and around users. But when we started doing that, the pressure increased. Many people fear the word pressure, but I want to say pressure has a cost and has a gift. What it did for us is bring up some of the dysfunctions, the areas that we could improve. During the last year, I don't know if you have followed the news about the wind industry, but we faced many headwinds. Capital became much more expensive, supply chain prices too. We also have regulatory and compliance issues, and also politics.

[00:07:41] The thing is that when growth is strong, many of these dysfunctions that we talk about can hide. But when growth is not as strong, that's when we need to react, and every decision matters more and more when you have pressure. So pressure did not create the transformation, but it exposed why we needed one. We saw familiar symptoms, and at the beginning we looked at the typical symptoms: slow delivery. What happens with a pull request? How fast are we delivering? How many features? Missed deadlines, product frictions, et cetera.

[00:08:21] We had many of these, and we thought that by solving those, and by asking questions like how do we plan better, how can we become more predictable, we could solve the issues, the symptoms that we had. But we wanted to look deeper, for the root causes. We knew that this was not enough. These delivery behaviors that we were seeing in the organization, these symptoms, were a response to the organizational design.

[00:08:54] If decision rights are not clear, the teams will stall or they wait. If we have too many dependencies and there is no transparency, you can work as much as you want with roadmaps, but it becomes an optimistic fiction. If we have governance that only focuses on reporting, then it becomes a bit of a theater, I would say. And if we had continued working in technology as an independent organization, far away from the business, far away from the users, we would not be adding enough value. Now Ather will talk about some of the strategic decisions that we have taken on the way.

Four strategic priorities and four responses

[00:09:39] Ather: As I was saying, when things were going fine, we were expanding in all kinds of areas. We were expanding in the US, mainland Europe, Asia, and also looking into the Australian territory. But then, since it became too expensive, Ørsted updated its fundamental priorities, and these were the four priorities that came from the top. The first was that we need to rethink our capital structure. For a long time the capital structure worked as a partnership model, where we had to go out and find another partner. When everything came under stress, it was also difficult, in one of the cases, to find that partner. So of course we had to look back, and what we did was go back to the market and raise the capital.

[00:10:36] The second one is that we still have to build close to 8.1 gigawatts of wind farms in the coming two years. In terms of context, in the last 35 years we have built close to 10 gigawatts of wind farms, and in the next two years we have to almost double that capacity. At the same time, we also need to see how we can improve on the geographies. There have been clear guidelines and guardrails, and that means that we cannot play everywhere. We have to be very selective about which areas, which industries and which continents we're going to play a major part in. And we are going to play a major part in mainland Europe, because that's where most of the value is generated.

[00:11:37] The last one is increasing competitiveness. When Ørsted started, they were one of the big ones, and we are also one of the biggest now; actually we are the biggest offshore wind company as of now. But we need to focus on how we can be more competitive, because there are lots of new players coming into the market wanting to try it out, and right now the market is quite tough.

[00:12:15] So, to see what kind of initiatives we could do, one of the big focuses we have right now is clarifying decision-making. When we kept on expanding everywhere, the issues within those decisions that I was talking about were not really surfacing. Now it has to be really clearly defined who's doing what, where exactly those cut points come, and what kind of collaboration model and governance needs to play an important role.

[00:12:53] The second one is organizing around value and not around technology stacks. In the expansion phase, the company expanded in many directions, but that expansion also meant that sometimes we expanded in a way that was not completely competitive. That's where we had to look into organizing around value, and I'm going to touch on that shortly. The third point is strengthening our craft. When we say strengthening our craft, we are mainly talking about two areas. The first one is the product management and product leadership craft, and the second is the delivery craft. Those are the two areas that we are supporting as an operating model cross-functional team. That's where we are focusing a lot more, to increase the craft of our roles within product and delivery.

[00:13:59] The last one is making governance a learning system. What that means is that we worked in a SAFe model from 2016 up to 2024. With that SAFe model we created standard structures across different areas, and perhaps those structures were not working. So we need to learn, change and update our approach so that we can get feedback from the governance model and then adapt and change.

Saying no and the decision stack

[00:14:33] One of the key points here is that Ørsted learned it the hard way. It was the market pressure that made us realize that we were doing some of those things not completely optimally. That also meant we had to learn to say no to some of the initiatives and activities that we had been doing for quite a long time. And one of the key points I would like to make is that when we work in this kind of context, clarity comes from deciding what we are not going to do. In the building industry, we are working with engineers. They are builders by design. That's the kind of mentality and mindset we have to train our different roles in: how can we say no to opportunities where we can see we can create value, if they're not connected to the overall strategy and vision?

[00:15:45] One of the key things we worked on in this transformation journey was that we built the transformation on Martin Eriksson's decision stack. How many of you know about the decision stack? Not many. There's a book that came out recently, a couple of months back, and the book itself is called Decision Stack. Basically, what Martin Eriksson is trying to help everyone understand is that you need to make sure that there's a coherent connection between the strategy and vision at the top level and the delivery that we are doing in different product teams, so that it doesn't deviate from the vision and strategy defined at the top layer.

[00:16:34] That helped our part of the organization quite a lot in defining the organization structure, and now we are on a journey to improve that: making sure that our roles, the product roles and the delivery roles, understand that they need to make their own strategy layer, their own decision stack within their own sub-organization, which can then be connected to the overall vision.

[00:17:13] In this whole process of transformation and designing the organization, one key point came up quite often, and that was: how do we define our product? What happened, at least in this organization, is that people often mistake a tool for a product. That's where we wanted to demystify it and clear heads, stating that a product is a digital tool, system or service that solves specific problems and needs for the users, and these are mainly our Ørsted colleagues, and in return creates value for the company. This is the key point we are still working on: if we are building a service, that service can also be called, and we're going to call it, a product. If we are delivering servers, different Wi-Fi updates to close to 1,800 wind turbines, how can we improve that service? Aitor, perhaps you can touch on the organization design.

Organizing around value, not technology

[00:18:22] Aitor: Yes. What I want to mention is that when we are designing the organization, the users don't care about your technology landscape. It is my experience, working in technology organizations, that most want to organize around a specific technology because they like it or because it's new. It might make sense from the inside, but the users don't care about that. They care about the user experience that they have. They are trying to complete a job. They are trying to operate. They are trying to manage a risk. They are trying to serve a customer. They are trying to move forward.

[00:19:04] For us there was a shift there. We were also organized pretty much around technology, but we introduced a shift in the mindset while doing this work, to organize around value. We started asking different questions. Instead of the delivery questions that I mentioned before, we started asking which of our digital products need to work together in order to serve the users, to serve the customers. By doing that, we came up with a design where we are grouping the different applications into products. We are empowering the product teams, so the product teams own the products; they have full accountability for them. Based on a product taxonomy, we group those product teams and products into product suites that serve a segment. And then a group of product suites forms what we call platforms.

[00:20:01] With that, we ended up with three technology platforms in Ørsted Digital Technology, and then four business-facing platforms, where we work closely with and are merged with the business. But it requires even more simplification. When you look at the organization, at the pressure that you have, at how the market is changing, you aim to simplify things. And boundaries add silos. Whether you want it or not, we need to add boundaries when we do organizational design, and with these boundaries you should be very conscious and intentional about where you want to add them, and perhaps those silos.

[00:20:51] My message to all of you is that whenever you are designing, or talking with the people who design your organization, look for where the value gets stuck. In this slide you can see some of the numbers from the transformation that we are going through. We have decreased the number of delivery organizations, some by grouping them, some by de-duplicating them, from 22 to seven. We have also looked at the supporting functions, which some other organizations call cross-functional teams or CoEs, and we have decreased those too.

A three-level governance model

[00:21:31] But these structures create time and space for better decisions. How can we ensure that we make better decisions? We do that by implementing a new governance model. We had a keynote before, I think it was Romain, where it was mentioned that the teams group together yearly. We have a governance that is split into three levels. In the first level of the governance, we look ahead at the strategy of the organization. Ørsted has a three-year strategy with strategic priorities that get updated. So in Digital Technology we meet, we analyze the strategy of the organization, and then we set the roadmaps and the plans. We have annual planning strategy days.

[00:22:24] In the middle, with a different cadence, quarterly, is where we align on delivery, and then we report, we look at the performance of the organization and we evaluate. And at the bottom is where the teams, quarterly and monthly, work together to deliver the products. You might notice that there are some arrows that go up and down, and this is intentional in the governance model that we have come up with. Our key principle here is that information flows from the first layer to the middle layer to the bottom, and from the bottom to the top. Teams need clarity from leaders, and leaders need reality from the teams. That is the principle we are following. And as I mentioned before, if this information doesn't flow, it becomes a theater.

Building the product craft

[00:23:18] Ather: Touching on the craft aspect. One of the key aspects within product management and product leadership was how we can build the craft. Many of our product owners and product managers have a heavy engineering background. That also means they are absolutely passionate about delivering, and often that's where it gets mismatched. The whole point here is how we can balance strategy and delivery, and how we can get insights, the product insights and the user insights, as a cornerstone to support our strategy but also delivery to our end customers.

[00:23:58] The role of a product manager or product owner is also quite important: his or her role is to balance strategy, delivery, insights and technology, so that we do not create silos across those roles. The product managers and product owners understand the technology stack, they understand the strategy, and whatever we are delivering, there's the right balance. That has been a big focus, and it's the focus we're going to have in the coming months.

[00:24:29] Going back to the point about the learning journey: right now we are on track to create a learning journey for our product owners and product managers. Soon we'll be doing some work on employing Martin Eriksson's decision stack in each of the platforms, and also building roadmaps together with those product managers, so that the roadmap is connected to the delivery that they are doing. These are the four or five areas that we quickly defined as the areas of the learning journey we need to have together with our product roles.

Where we are and lessons for Monday

[00:25:08] Aitor: Great. As we mentioned at the beginning, we are still on the journey. We are not done. We have taken some steps: we have simplified the organization, we have established the platforms, and we have grouped the teams into product teams and product suites. Now, thanks to all the work that we have done, we are evolving our delivery model. And we still need to focus on many areas, like the decision stack. We need to focus on performance, which has been mentioned today. We still need to focus on establishing better collaboration models, and also on the culture in the organization, as has been mentioned by Marcus. So let's close with some learnings that you can take for Monday.

[00:25:56] Ather: Yeah. In short, there are four points that I have put together. The first one you can take with you is that organization design is not separate from the product work. If the product is broken, perhaps there's also a broken system behind it. That's something you need to ask about and look into. The second one is that under pressure, the system reveals itself. We enjoyed a really nice ride for many years, but then we came under pressure, and that's where we really hit reality. So I think it's a good idea to start being a lot more transparent about the priorities and the work that people are doing.

[00:26:36] The third point is the autonomy part. For a long period of time, within the SAFe Scaled Agile Framework model, we focused on autonomy, autonomy and autonomy. But that autonomy approach also meant that each team was working on its own. So how can we build a system that supports that autonomy but is also connected? That's where the decision stack helped us. And the final one is that product leadership is not just a role. If you are a really great product manager or product leader, that's great. But that product manager or product leader needs to fit into a system, and the system should support them, and there should be some kind of coordination across. Thank you very much.

[00:27:22] Aitor: Thank you very much.

Q&A

[00:27:29] Host: Thank you. That's brilliant. I'm one of those nerds who loves organization design, so I could probably keep asking questions about that all day long, but we'll hand over to our audience to ask the questions first. What is the impact on Ørsted of the changed US energy policies? Wow, starting nice and easy.

[00:27:49] Aitor: I think that's a very good question, and I will actually take perhaps the easy answer here. It's very challenging. And to be fair to the organization, I think it's fair to say that to know more, you can Google it, because we cannot disclose confidential information. But it's very challenging, we know that. It's also challenging in Europe and other areas, but much more in the US.

[00:28:14] Host: Okay, and who knows, policies change over time, so I'm sure there'll be updates in the future. What is the biggest learning about org design that you wish you had known prior to this, and did you have to be an octopus to do it all?

[00:28:31] Aitor: Well, I don't know exactly what the octopus way is.

[00:28:34] Host: Eight arms, I guess.

[00:28:36] Aitor: One of the biggest learnings about organizational design is perhaps: talk to your users, as we talked about here. Talk to your employees. Ask where the bottlenecks are, what is impeding us from delivering, what is impeding us from collaborating together better. And also talk to the users and see if you can find out, by talking with them, where there could be some systemic problems. We heard a lot in the conference about systemic problems. Systems thinking is super, super important, and I know octopuses are very, very intelligent animals. So we need to be like them. We need to be open and we need to listen from different angles.

[00:29:22] Host: In one of your slides you had the reduction, I can't remember the exact numbers, in the number of units and centers of excellence. Having tried this kind of organizational change, there are a lot of people who had an organizational unit who no longer have one, so there was a huge political pushback. Did you face that, and how did you overcome it? Because you're saying to somebody, you've got a unit, and now you don't.

[00:29:47] Aitor: Yeah, I think we faced that. It would be unfair to say that we haven't. One of the things that helped us during this transformation is that the mandate came from the top, from the executive team. And we know that we are under pressure, and that makes everybody in the organization somewhat aware that we need to react, that we need to do something to continue being in the business and being as good as we have been so far.

[00:30:19] Host: Excellent. You said we should organize around value instead of technology, which I tend to agree with. But how do you fight the silos that kind of organization creates? We're shifting from a technology silo to a value silo. Have we just swapped one silo for another, and how do you keep knowledge sharing across?

[00:30:39] Ather: I think we cannot completely remove those silos, I completely agree, but to some extent we can. I'll give you a concrete example. Say we have five different products that we are building in different silos around inspection. There's inspection in the commissioning phase, there's inspection in the operation phase, and there's also inspection in the building phase. If we now have the user persona, which is around the inspection itself and the person who is going to inspect, then to some extent you are going to remove those silos. We are not going to build five different technologies; we're going to focus on one technology. But you cannot completely remove all those silos. Yeah, that's hard.

[00:31:26] Host: Okay, we'll have to be quick with this last question. You mentioned a book about the ability to decide. Can you repeat the title? Brilliant quick question.

[00:31:35] Ather: It's very, very easy. It's from Martin Eriksson. It's called The Decision Stack.

[00:31:40] Host: The Decision Stack. Perfect. Well, that actually brings us to time. Excellent. And Martin, you can pay us later.

[00:31:50] Aitor: We are not a sponsor.

[00:31:52] Ather: No, we are not.

[00:31:53] Host: But that brings us to time. So thank you very much. Thank you, Ather.