 Mae David i wnaeth i chi'n cyflwyno how organisations can become digital enterprises So plis a warm open group welcome for David Weeble Thank you It's just as good the mic's on So, when I was asked to speak here I was actually quite happy because the topic was actually being a digital enterprise not about digital transformation And hopefully some of the content I'm going to talk about today will actually put that into context Ieithio rhoiIEI yn dweud i y gallu gwanful o byd. Mae gennym eich gweithredu, roedden nhon yn y prônysau. Mae'r gweithredu yw eich gweithredu yw'r gwanful. Rwy'n iawn i fynd i rydych chi arnynedd i妹ol, mae'r bwysig y context. Mae'r context eich gwaith i fynd i gweithredu. Y gallwch y prynidol ychydig yng Nghymru? Mae yna gallwn rhai bod ddach i ddefnyddio'r bwynt hynny. Eu pad yn amser mynd gan fel fod eich dweithredu ni'n gwybod gwerthu'n eu gwaith. I'm actually going to talk about it from Forrester's point of view on what being a digital enterprise really means. And then finally I'm going to sort of start talking about some of the skills and the changes that are necessary for actually being a technology group in that sort of new organisation. So to start with, Forrester works across both business and technology companies, leaders and people. So we don't actually look purely at technology. We are actually doing across the business as well. Now we recognise that sort of the empowered customers are now creating and remaking markets and creating what we're saying is actually existential risk for companies. So customers are making the rules now. The way I describe it is in the past generally businesses made the rules. Businesses built processes of how their products and services would work and then expected customers to follow them. Now the customers tell you how they want to interact with you and you have to go with how the customers want. Digital is destroying industry lines and business models. So traditional business models no longer work. If transformation does not occur those businesses can fail. And equally digital platforms are actually creating a barrier between brands and their customers. This is an interesting point because several people asked and said actually surely it's getting us closer to our customers. But what we're finding is the sort of the mega brands the Amazons and so on are actually the ones that are creating the contact with the customers. So other brands are being separated. And I was talking to someone actually from the BBC a few weeks ago and he said this is happening to them is news is no longer associated with the BBC. It's delivered by another platform and no one knows where it comes from. And companies are struggling to become digital first. So that actual switch over to being digital is taking is a huge amount of effort which is obviously part of the conversation for the conference this week. And then the gap finally the gap is widening between those leaders and those incrementing forward slowly. So this is not a great scene if you're thinking about traditional business and thinking about the changes that are necessary. These are lots of challenges that are occurring for organizations. So what does that mean? Well actually means that business and technology leaders are actually making decisions now that will decide the destiny of their companies. Now we may say that that's quite a harsh statement or difficult statement. But if you look at the impact in say the retail sector at the moment with companies struggling to adapt and change and many it's particularly in the UK at the moment going out of business. The comment I made is that it's easy to assume that they're actually not being run very well that they're bad businesses. But these businesses have existed for many tens of years sometimes hundreds of years. They're not bad businesses. They're just struggling to adapt. So we're saying that these powerful external forces are actually requiring immediate action by by organizations. Decisions are being made with little precedents but massive consequence. So again the typical behavior in the past has been for looking to see what's happening on the market. See what other companies have done. How have they changed. How have they succeeded. But now that's not working any law. So decisions have to be made but you can't look to the market to say whether they are successful or not. And the decisions are not cosmetic. They actually go to the heart and the very nature of companies. So seeing interpreting and acting on those external forces is now becoming a critical capability. So what does this mean? So Forrester talks about the age of the customer. And some of you have seen me present about this before. I'm not going to go into what that means in quite so much detail. But essentially it means putting the customer at the heart of everything you do. So we say that there are a number of principles that mean that companies must adapt to or use. And the first is actually being customer led. Not being process product led but actually understanding what your customers want and how they want to be interacted with. Insights driven. Using data to create and drive strategic advantage. Fast. Rapidly respond to changing markets and market disruption. And then connected. Orchestrating internal operations and ecosystems. Now I'm going to talk about these as I go for the presentation and come back to sort of the drivers for what we're seeing as the way forward. So we use these actually in terms of what we call the customer obsessed operating model. Now this is not a pure technology model. This is actually looking at what we're saying to organisations and businesses overall. So that customer led insights driven fast and connected. And our definition, Forrester likes definitions. A customer obsessed enterprise focuses on its strategy, operations and budget to enhance its knowledge and engagement with customers. Now within there we see this there's sort of six levels. Technology, structure, culture, talent, metrics and processes. And I'm going to talk about them. But one thing I just want to sort of highlight. This might be a European thing so I apologise. We use the term customer obsessed. Not customer centric, not customer interested, customer obsessed. In Europe obsession perhaps has a negative sort of meaning as well. But what we mean here really is the fact that everyone in the organisation understands how they impact the customer. So if you're talking to the technology department, they must understand how they impact that customer experience and that customer interaction. So when we're looking through what we truly say truly customer obsessed organisations, it is going through the whole organisation. Everyone understands what the customer means. So if we look now into that transformation piece, what does that look like for these modern organisations? So a traditional transformation, the phases of transformation are generally sequential. They focus on internals, decisions are made on sort of direct experience or gut instinct and activities are labour intensive. If we translate that into what we see successful digital transformations looking like, then customers are the starting point, not the internal. Decisions are based on data, not on instinct or previous experience. You're looking for accelerators to shorten and improve processes and phases are agile and iterative and overlapping. So looking at those metrics and the levers that we could talk about, how does that relate then to what we see actually going on in the market? And so this is where some of the updated research comes in. And this is where it paints a stark picture of actually what is going on in sort of the world of organisations. So clear thing is every business is becoming digital. There is no industry sector that is immune to this. It doesn't matter if you're a B2B, B2C or any combinations of those, every business is effectively becoming digital. So we as Forrester, we survey large numbers of businesses and this is a global survey. And we ask them how many are actually doing a digital transformation. As you can see 58% currently underway, 14% investigating, 6% improving but not transforming and 22% say they've completed it. Well, what's interesting about that is if 22% think they've completed it, they've really not understood this transformation or being a digital enterprise. Because the world is still changing. So how are they actually managing to say they've completed their transformation? So that lack of understanding about what the transformation actually occurs, it's still out there, what it means to be digital. So from the data we then see three patterns emerging. We've categorised those organisations then into the digital beginners, the digital intermediates and the digital advance. And for us we're really saying the beginners are looking to disrupt experiences. The intermediates are disrupting their businesses and they advance the disrupting markets. Putting that into sort of what that means, so the beginners then are just looking at building new experiences. They're adding digital to existing products. The intermediate firms are the real, where majority of the people are. They're actually transforming their business from the outside in. They're looking at that customer, looking at the customer journey, understanding where they add value. And then the advanced firms are actually looking at new ways to deliver those outcomes to those customers. And so what we can say then is the beginners really support existing models. The intermediates innovate products and services and drive new products and services within that model. And then the advanced are the ones that are actually changing their operating models, embracing digital ecosystems and building new sources of revenue. Now actually going through the data again, the difference between those three levels is quite stark. Now this is the good news slide because this is actually one where a few years ago it didn't look so good. Now at least we say that every company has a digital strategy and the majority of the execs actually understand it. When we went back a couple of years ago, it was clear that majority of the execs did not understand their own digital strategy. At the beginner level you still say there's 24% that don't understand it, but it's a great improvement. But looking at those levels of the customer obsessed operating model, then we see a sort of more stark picture. So looking at the culture, only 12% of organisations at the beginner level think they have the culture to succeed in their transformation compared to 94% at the advanced level. Looking at the technology, only 15% think they have the technology to achieve their digital transformation, whereas again 89% at the advanced level. From the processes side, only 5% of the beginners think they have the right processes to be able to succeed at digital compared to 84% at the advanced level. And then for the talent, you can see where the right people or the right skills and people, it becomes very, very, sorry I hit the wrong button there, very, very sort of difference on there. So looking when I said at that customer obsessed operating model and then looking at the difference of people on their transformation, you can see this is where we're saying that those advanced companies are now accelerating away from those that are stuck at the early stage of the transformation. This is the real challenge that we sort of face now as organisations in transforming and becoming digital enterprises because those advanced ones are really are the digital enterprises and everyone is now playing that sort of catch up. So what does it mean to be a digital enterprise? It's already good to talk about it in abstract but actually how do I apply that, what do I need to look at? So from how Forrester looks at the market and what organisations have done, so again from us we're trying to analyse what companies are doing and what are they looking at. So we look at two different sides of capabilities to being a digital organisation. One is the digital customer experience and the other is the digital operational excellence. So what we're talking about in customer experience terms is digitising the end to end customer experience, understanding what that customer journey is, understanding what the digital touch points are as well as the traditional touch points and making that way enhancing it through sort of digital as well. Enhancing those products and services as part of the customer's value ecosystem and I'm going to talk about what I mean by value ecosystem a bit later on. But finally, if we're also going to actually build these digital connections, these digital sort of interactions, our customers need to trust that we're doing this in the right way. So we have to create trusted machines. Are we going to protect data? Are we going to actually use the information we gather in a ethical and a correct way? On the operational experience side, we also look to digitise for agility over efficiency. That is a big turnaround for many organisations because most of automation until now has been done for efficiency and not for agility. Drive rapid customer centric collaboration innovation. How do you actually get that fast, the speed to actually change things quickly, be able to adapt to changing customer needs and expectations more rapidly and orchestrating new capabilities to rapidly enhanced customer value? What we mean by this is that you are part of an ecosystem and you need to orchestrate internal and external services to be able to meet those customer needs. Again, I'm going to show a bit more about what I mean by that. So I mentioned ecosystems several times. What do I actually mean by ecosystems? Well, this actually comes down to the reality of how people now live their lives and work. So this is the one I use quite often because it's obvious. So my desire to have a productive business trip. So I travelled here from the UK to get here. I needed to book a flight. I needed to check which seats I'm in. Am I happy with that? Booked a taxi, hotel, got to the airport. I needed to find somewhere to eat. I needed to find my way around. Now to actually do that, I actually built an ecosystem of apps that I used to actually do that. There was no one single place I went to. So what I'm doing now is I'm putting together an ecosystem of technology, of digital services that I use to achieve an outcome. And that's what we mean by these customer value ecosystems because how your customers will interact with you is in a similar way. They won't just use your products. They'll use a combination of services to achieve their outcomes. Now that is occurring both on the customer side but it's also occurring on the business side. So businesses are no longer just isolated and control their own technology and their own sort of processes. They're actually forming ecosystems now. Now particularly in Europe in the financial services where a lot of regulation is driven open standards, we're really seeing this start coming to the fore. So if you look at say small business services, their ecosystems now cover having an account, doing payments, doing accounting. And what we're finding is these are becoming digitally integrated to offer consistent service to those clients across multiple providers and multiple products and services. So how do we look at that and where do we start from a digital perspective? If you remember those numbers on transformation, well if I look at the digital experiences one axis and the operational excellence another axis, then we get the sort of the classic consulting analyst four box model. So many many people are still in the digital dinosaur area. They're not looking at the digital customer experience and they're not operating on digital operational excellence. Those that are focused on efficiency have generally become the digital operators offering great customer cost savings, cheap payers, but not differentiating on customer experience. Those on the other hand of concentrating on the customer experience but not the operations have come digital connectors where they're actually driving a level of customer experience. But those that do both are the ones that are becoming the digital masters and we're saying they're the predators in the marketplace because they understand it. And we see examples in all of those. So last year we're doing some work with a company, a catering company who do large scale catering. They had one of the best run IT departments that I've seen. They knew exactly what they were doing. They had very very good control, very very good governance. But when we got them to actually assess what they were doing they assessed themselves as digital dinosaurs because they couldn't actually respond to this new world. And even though they were running IT very well they were not actually responding to the business very very well. And so they started using these frameworks to help them understand how they had to change. So don't think again digital dinosaurs means that things are being badly run. It doesn't mean it's at all. So where does innovation occur? Well actually innovation occurs between those two axes. The operational excellence and the customer experience. And it becomes a cycle because as you improve the operational excellence you can in general improve the customer experience. And that's where we find those why those people going on one axis and not the other are not becoming the masters is because you can't do one without the other. And we're seeing those companies that have gone along one axis now having to move along the other axis as well. So question then to take this forward is how do we manage the future? What does this mean particularly for technology professionals? So the challenge, the usual perception is that change is actually linear. Change isn't linear. That's a real thing. When you look at change and how it occurs it actually is exponential. So taking a quote from the law of accelerating returns. So we won't experience 100 years progress in the 21st century. It will be more like 20,000 years of progress at today's rate. And if you look at how the different levels of technology have built on themselves going from chip storage through to cloud big data, now through to blockchain, quantum, AI, Edge and IoT we're seeing this ever, ever accelerating build of technology. And it's not slowing down. So if I actually say looking at the emerging tech that we analyse and just start putting it onto a timeline. World time interaction management for systems, IoT, intelligent agents, AR and VR, personal data. Look at systems of insight, AI, cognitive, spatial, insight platforms, IoT analytics and then look at the systems of support. So from today with cloud native platforms, containers all the way up through to hybrid wireless and Edge computing. Now there's only one thing I can say really about this as a roadmap. I can guarantee 100% it's wrong. So it's a good idea. It looks good. However, this will not be reality. And it doesn't matter which order I put things on in there. It will not be reality. So we've got to deal with this change coming up when we can't predict what's going to work and what's not going to work. Now I put this slide up. This is a slide that is common from many of my presentations. The reason I put it up is it's only three to four years old this slide. And look how dated it looks. So the technology in there already does not look like it's what we expect as consumers to be for modern technology. So again, that speed is happening. One of the CIOs I work with, Tim Hines from AIB Bank in Ireland has a very good way of describing this. He says that we tend to overestimate the level of change that's going to occur in the two to three year period but underestimate the amount of change in the ten year period. If you think about it, if you go back in ten year increments the amount of change has been phenomenal and I said that's not slowing down, that's increasing. So again, what does it mean for technology professionals? Well, the clear indication for organisations is they don't think current IT organisations can meet this challenge. So when we surveyed and asked here, and you can see this is 18,000 business technology decision makers, a large number of them are reorganising their IT departments to do this. So they're feeling the current IT structures, current IT organisations need to change to be able to support this new way of working. So for us, what does this mean? What it means, IT success is not measured about how well it responds to the business but now about how well it enables us to work within this environment and compete within this environment. And that's an important discussion point because earlier on when we were in the foyer before we actually had a conversation about business relationship management and how to work with the business. So we've got to change our view. It's not so much about business alignment. It is actually about business enablement and how does the business relationship managers, how do those people work and how do they enable the business to compete in this new world. So we're saying for CIOs the priorities have changed. These are the five things that we say they've got a key capabilities that they have to concentrate on. Flexible sourcing, information security, customer and strategy and design, continuous planning and governance and continuous delivery. Flexible sourcing. That's an interesting one to start with. So one of the key things we're saying here is stop writing RFPs. They don't work. It's, it wastes a lot of people's time. It doesn't do you any favours. It doesn't do the vendors and suppliers any favours. Reason being is the recognition internally in development is we're moving towards agile, to DevOps, to adapting. So how do you think you can write an RFP with hundreds of requirements that are going to tie a supplier in to five to ten years? It's just not going to work. So flexible sourcing has got to be about variable resources, continuous governance, response to shifting demands. A true partnership with those suppliers are much more flexible in its structure, not these fixed style contracts. Information security, zero trust approach. So really saying that security has to be embedded into every element. You no longer have a firewall fortress around your organisation. So every element must have security built in. Customer in strategy and design. So really actually trying to do true design with customers, understanding what the customer's perspective is, not what you think it is, but what their real perspective is. Continuous planning and governance, innovation, continuous planning, not just on budget cycles, but how do you continually look at your portfolio and review and update and respond as necessary. And then continuous delivery, DevOps, agile, business engagement, continuous testing. All really about that ongoing and continual way of working. So what does that mean in terms of what we actually see at the moment, some of the changes that are necessary if you look at those five capabilities? Well actually unfortunately where we spend most of our time at the moment is on those operational value streams. So using IT for IT as the sort of framework, detector correct, request a fill. Yeah we're good at those, we know generally how they work. What we're not so good at is the strategy to portfolio, the requirement to deployment. And if you look at those five capabilities I said, lots of them fall into those sort of strategic areas, and those areas which we need to develop and take forward. And that is a big shift again in skills and capabilities that people need to make. So now take this into what does it mean for sort of architectures and how we work and the sort of challenges around there. So this now means this is a sort of modern approach to application development. We build microservices to sort of componentise and decouple the applications that we deliver. So instead of delivering a monolithic application that we've tested together, what we've done now is build all these microservices which we deliver independently. And those microservice may be built by separate teams depending on their capability. So this is sort of where we sit. Now one thing I try to point out here is that if you look at that and I took away any of the stuff that said microservice, it would still have the same complexity as a traditional application. So we've not removed complexity here at all. We've just changed the way we build and deliver that application. Now unfortunately that complexity is not going away, has certain impacts. So let's take a sort of step forward now and look to the future. So remember I talked about ecosystems. I talked about how I build a personal ecosystem. So going back through that, what does that really look like in there? So if I want to go to a trip to San Francisco for example, I have all these questions that I want to answer. So just as I described from my trip here, all those apps that come together and all those apps have different APIs and different approaches. So we've now got to start bringing those components together if we're talking about that complexity in microservice delivery. But the problem being is because we haven't got rid of the complexity, we've just moved it. So we've moved the complexity from build time where we treated it as a monolithic application, we treated it, we worked with it to make sure it all worked together and then we delivered it. Now the customer is getting all these bits and pieces and that integration is occurring at runtime and often with the customer combining components that they're selecting. Now I've seen many people talk about automation in what we do in IT, not necessarily the business. And they talk about it in just process automation and there. So being clear here, I'm not talking about process automation, I'm talking about our ability to manage these components. So if you look at it, manual ability actually is fairly linear. You can only add people so often to get the scale. So if we're struggling at this stage of the sort of life scope, how are we going to move forward? And that's where automation is going to have to come in. We're going to have to have much more automation of those environments and how we deliver those experiences. So for us, for a definition here of what we mean by sort of that digital business technology environment, we talk about digital business platforms and built of modular technology built around business APIs designed for rapid reconfiguration of business models, processes and ecosystems. So this is really what we're aiming for in terms of meeting those requirements in the model world, of those business demands. We must have a system that is actually capable of being resilient to absorb high amounts of change and that change can occur at runtime, not necessarily at build time. Now one of the other changes that that implies is the other thing, insights driven organisations. What does it really mean to be an insight driven organisation? Well here, I have to be careful and say it clearly so I'm not misunderstood. What normally occurs is that people assume that means more business reports, more reports and read and decisions. Actually what insights driven really means is deploying models that actually respond to the data. So if you look at this data sites lifecycle, we understand the data, we prepare data, we build models, we evaluate that model so it works, but then we deploy it and that model will then drive that business action. So whether it's optimising supply chain or optimising customer interaction, that's how the model now works and we see AI and machine learning now coming into this. A common mistake people also make is assuming that this is going to be understandable and sort of human like intelligence. Many of these models are not understandable by humans so one of the companies in Germany that optimised their supply chain actually had a deep learning neural network that analysed 200 different variables to make its decisions. No human can collo relate across 200 variables so these models don't necessarily have to be human understandable and that's where we're seeing some of the optimisation coming in. So where does that drive to all this experience, the components that are coming together? So today if I describe that customer experience the way I worked is we go from our needs, I create that ecosystem of value by translation and it drives a series of actions. We're already getting to the point where my needs and the translation into actions are now being automated by some of the different applications we see. Such as on my phone I have Android it now does a lot of predictive of which apps I want to use and that future state is where the predictive analytics and models now come in to actually predict what I'm going to do. If you think this is not actually real and look at just Amazon Echo it's not just the ultimate relationship platform but actually it's the ultimate service integration platform. It's a virtual assistant, AI system that sits in front of 20,000 plus microservices which I don't care how it works. I just ask it a question and it finds out and answers me the question or brings in the relevant service for me to be able to achieve that outcome. So these systems are there, they're real and it's not pure future we're looking at. So how does that mean when we work? How do we work? Well again we've got a lot of focus on DevOps and agile and things like that but in some ways the business is responding and realizing that they have to work in a similar way. So we've seen the rise of design thinking as a business approach. Design thinking really just comes to the same realization that IT has come for in that we don't really know what the customer wants and pretend that we can write it down is wrong. So we have to actually experiment. If you actually map the sort of phases of design thinking then on to a life cycle from the empathize, define, IDA prototype, test, analyze you can see there's a range of capabilities here from research to personas to customer journey mapping through to agile development DevOps, instrumentation, usage and blue-green type of deployments. Now what I normally point out here is there's no business and technology divide now on this slide. There is no point in which we're saying that's business, that's technology. It's now much more of a continuous approach on how we adapt to it. And one of the key things about here is each of those stages actually forms parts of divergent and convergent thinking. So empathize is divergent, define is convergent, IDA divergent, prototype convergent. So we're actually looking at this way of working which is changing, but it's an integrated and continuous cycle. So finishing up one of the things I like to point out as a quote from Sapiens is that the scientific revolution was not a revolution of knowledge but actually it was a revolution of ignorance. The great thing that drove the scientific revolution was actually the recognition that we didn't know the answers to the most important questions. And that's the similar thing that we have to now sort of do with digital and being a digital enterprise. Because what we need to understand is we don't know all the answers. So lots of these approaches now around testing, learning, and actually learning about what the customer wants. So I sort of half joke but half serious say if you're willing to do this fail fast, well fail fast is good but actually what you need to do is learn. If you fail fast or even succeed fast you don't learn anything, it's not really given you value. So the important part here is what did you learn from the experience. And that's where the fast comes in, the ability to experiment, the ability to learn and gain knowledge. So if you're implementing DevOps and you think let's try it out on the simplest customer app, it's not going to teach you anything, we know it's going to work. All you've proved is that everyone else has proved it works. You actually have to try some of these techniques out on the hard problems. What's it really going to teach you? What's it really going to learn? And with that, I'm hopefully just about on time. I will take some questions from Steve. Thank you very much, David. Please have a seat. Thank you very much. Interesting as usual. We've got a number of questions that have come through on Slido. It's not too late at this point but it soon will be for those of you who haven't. The first one that came in was you were talking about certain you're seeing organisations digitised for agility, not just efficiency. Can you give examples of what you're seeing that's different? One of the hard parts about where technology and IT has come from at the moment is this whole focus on efficiency. If we're talking to organisations, I still see it so often, cost, cost, cost, we're going to reduce cost. That isn't always the great way of looking at it. For one side, most organisations actually don't really understand all their costs very well. The other side is because of the financial pressures, many have been under investing and they've got a large technical debt built up. Those companies that are succeeding are the ones that are really going back to the fundamentals and understanding where does value come from in their organisations and linking it back into the customer journey. Realising that the change and adding value to customer at agility then gives them advantage and therefore business value. Those that are still focused on the efficiency aspect are generally the ones that are changing too slowly and not being able to adapt. Just going on one of the axes that you were talking about. Question for a minute. That was for me, I'll answer that in a minute. How are the digital enterprise challenges affecting B2B organisations? Can you give any examples? Yes, we say B2B, B2C, there is very little difference that we see certainly in the disruption. I get this quite often. We're a B2B organisation, so customer experience doesn't matter to us. Simple answer is, are there humans still involved? If there are humans still involved, customer experience does matter. Even employee experience, what the employees experience, what their expectations are matters in there. That human element is still very, very important. All the things I talked about from a B2B or a B2C point of view are still valid and relevant and we're seeing that more and more come to the fore now in organisations. Everything really from very, very old conservative businesses all the way through to what we think is the modern ones. I wouldn't say it's one of those odd situations where I've not seen one industry that is not being impacted by these changes at the moment. OK, thank you. When do you think that digital transformation is complete or is it just a continuous process of learning and adopting? Actually, so I take to the second one, is if you get to the point where you have an organisation that is almost continually transforming, continually looking at how it can serve its customers, then that's when almost the transformation is complete but it's not because you're continuing to transform. So it's a cultural thing. It's the understanding that you have to adapt and change. So we can see some of the organisations, the ones that are obvious ones like Airbnb, they apply design thinking at every stage. So if they're starting a big new project, they'll look at their organisation for that project using design thinking. They will constantly review what is it that they can do better and how they can do it. So there isn't an end point to it. It's an ongoing, but from a... Why I like the title of this one is because that's being a digital enterprise. It's no longer really focused on transformation because unfortunately transformation now just means so much to different people. It's really that mindset, the culture, that then drives those organisations forward. All right, thank you. You mentioned no more RFPs. Partnership requires even more definition of the boundaries, expectations and responsibilities. How do you do that if not in an RFP? I was hoping that question would come up because that's usually one of the controversial ones. So it comes down to if you're working now within ecosystems and the boundaries of your business are not so set and your value as part of that ecosystem, then you're going to have to create partnerships or at least working relationships across there. Those working relationships can take a number of forms. But if I take the simple one, is look at say vendor tiering. Typically vendors or procurement departments only tier vendors based on sort of cost at the moment. What we're saying is you really now need to understand what you're trying to achieve from that. So yes, you'll still have your sort of tactical suppliers, staff augmentation, but they really are managed on cost. You may have then more strategic partnerships with companies that can actually help you win. Those partnerships have to be defined on value, but it has to be value for both sides. So it's no good if you, as a customer, win and make lots of money, but your vendor or supplier does not make money out of the deal. So what we have to have is shared outcomes, shared, actually shared vision of what that means to be a true partnership. And you said you're no longer running that partnership on cost. The extreme end is then say with your niche sort of suppliers, those that really drive high business out value, but also potentially high risk. So Fintech, again, is a good example of this. Many Fintechs are very small startup companies. They would not be your traditional suppliers. These companies may not even really be run very well because they don't have the experience. So we're seeing many banks actually having to step in and almost support those Fintechs to drive the sort of commercial advantage. So in that new world, actually it's not about writing lots of requirements and expecting the vendor or the supplier to be able to stick to them. It's understanding what relationship you want to have with them and how do you define that relationship so it actually works for both sides. All right. Thank you. We've had a flood of questions come in. We won't be able to get to all of them, but let me pick. To be data driven across an ecosystem demands a common understanding of the data across the ecosystem. How do successful businesses achieve this? That's also a very common question at the moment. It's what we're sort of saying, so from an architecture point of view and then from a business point of view. From an architecture point of view, people have got to have access to the data. Data is going to be the most important commodity going forward. Being sort of blunt about it, the algorithms and the technology are commodity. You can already buy machine learning algorithms. You just go on, you know, TensorFlow, whatever. But to work well, they need good data. If you don't have good data, then you can't actually do anything with that technology. So data is going to become the main component of any of the drivers of this. That has to work across everything. It has to be clear, transparent, but you have to govern it correctly. You've got to create an architecture that allows this and we talk about information fabrics, we talk about hub and spoke architectures and it comprises a very complex technology stack at the moment that's still maturing. But above that, you're then going to have the organization. So, you know, the technology department does not own the data. The data is still owned by business owners. But they've got to understand their role in governance, curating data, actually being able to self-serve data and use data. The final part there is, is we still see lots of organizations think the way to do this is go and hire lots of data scientists. Well, what we're finding is they go and hire lots of data scientists, after six months get bored because they haven't got the data, and then get a next job somewhere else because the pay is still going up. So actually understanding how you curate and supply the data to those groups, how you enable those data scientists to build the models so they can be deployed, so having the architecture that they can then respond to those, that sort of data in the correct way and having the real-time feeds. So taking that example of the German retailer, they have a model said with 200 variables that updups their supply chain, and it's taken 40% of cost out of their supply chain. But they have to have the real-time data coming into that model to then be able to, for it to work. And also being clear is that model replaces those typical supply chain people. So again, whereas ordering of the supply chain used to be really gut experience, now it's truly data driven and it's truly occurring automatically based on this model that's been trained. Thank you. The last question in the interest of time. You mentioned data scientists, so you've kind of played into it. We've heard for some time that they're much in demand, and in fact next time we have an open group quarterly event we're going to be able to announce a certification programme for data scientists. What other skills do you see either in great demand or short supply, or therefore probably both? I'd say generally the range of skills is overall, from end to end, is in demand. So the complexity, the rate of change means that the job is not getting simpler. But starting with your one, data scientists obviously has won this whole ability. So where the insights driven, the model led now, organisations are getting such competitive advantage. The business leaders see this as well. So they're saying data is our sort of key asset or we're going to monetise data. What they haven't got then is then the architecture to support that or the level of customer knowledge to be able to drive that. So this is where we're seeing those laggards really falling behind. They just do not have the capability. So that whole information area from the technical architecture to them to be able to build appropriate models is also in demand. But then actually looking at, say, the strategy roles, how do we create the right strategy? How do we understand the emerging technologies that may impact or may not impact? That ability to be able to experiment, so actually having skills around design thinking, not just DevOps, but understanding customer journey, so customer experience knowledge. So I was actually talking to a head of CX at one of the UK banks a few weeks ago. And with her, she seemed to think everyone understood CX. And I said, actually, if I go and talk to most technical people, I said, do you think customer experience is important? Lots of people put their hands up. Most people put their hands up. And I said, do you understand customer experience and then very few people really do? So again, it's those sort of skills that now about understanding the customer, how you design for the customer and not how you design pure just technology solutions. David, we'll leave it there. I'm sure you'll be approached in the break by people whose questions didn't get asked. But thank you once again.