David Boyle at Publishing Innivation Summit 2017.mov

Yeah, Some of the top we've talked about today. Loved the session earlier on, like ad level optimization and message optimization. There's some really helpful tips and tricks in there, but we step back a little bit to think instead about kind of corporate level optimization or brand level optimization, maybe. So the other sessions then were about what are these new technologies and tools enable us to do in terms of ad targeting, ad optimization, and I'm going to answer basically the same question, but again at the kind of brand or the corporate level, what do all these new technologies mean and what do they enable us to do differently to manage a corporation or to manage a brand in this new world? So, stepping back, I work at BBC Worldwide, so we're the global and commercial wing of the BBC, the British broadcaster who was excited yesterday about the news for Doctor who. Yes. I'm glad somebody cheered. If you don't know what that is, you're going to have to look it up later. Very exciting news. So Doctor who is one of our big global TV shows. 54 years of fantastic science fiction. Sherlock. Who watches Sherlock? Or seen that? Thank you. Less enthusiastic than the Doctor who, but all right, thank you. And anyone like Fast Cars, Top Gear. Awesome. There we go. And then Natural History content we're going to talk about later as well. Lots of great British TV content. Our question is, how do we help this content that's been produced in the UK by the BBC? How do we help this define global audiences? So certainly ad optimization and message optimization as part of that, but we need to navigate this complex world that's being built. We need to navigate all these new technologies, all these new routes to market. So I want to talk about what that feels like, what it looks like at a corporate or a brand level, and then again, some of the tips and tricks for how we navigate it. In particular, how we use technology to navigate it, with a focus on data analytics, insight, strategy, that kind of thing. So, four parts, really. First, like, what is all this innovation that we're seeing? How does that feel as a media company at a high level? Second, obviously, that's bringing complexity to our world. Massive, massive complexity. So what does that feel like? What does that mean for our business? But what does it mean for how we use that data and technology to better understand our opportunities? Understanding. Then we move on to. And what do we mean by understanding? And how do we build it? How do we bake it into the organization? How do we help people to better understand the world and make better decisions? And then focusing really on one key recommendation to come Back to the title of the presentation. One key lesson that embeds in everything that we do helps us to not just stay on top of this rapidly changing world, but to thrive in that environment. So first on the topic of innovation, then, I mean, unprecedented opportunities for media with all the innovation that's out there. So whereas it used to take what, 75 years for the telephone to reach 50 million households and Pokemon Go does that in just 10 days. So rapid, rapid increase in how quickly products can reach audiences globally and therefore tons of new exciting opportunities in the media world to reach audiences. I'll show you what that looks like in the TV space. I think you can do a parallel version of this slide for retail, for consumer goods, for any industry you want. It's the same basic story and it's scary and exciting at the same time. If you rewind back in time, some of the executives at the BBC will remember a time like this, a very simple world. Our product went to market via television and some people had computers. Very, very simple world. And of course, both of those things got infinitely more complex relatively quickly. The television. Now there's so many different ways video content reaches consumers via television on demand, custom navigations, et cetera. The computer was added to with a phone, but a relatively simple phone. Again introducing complexity, but still a kind of manageable amount. And in recent times, each of those has broken into loads of different individual platforms, loads of individual routes to market, loads of individual product opportunities available. Starting to get hard for any one person to really understand all the ways content reaches consumers and all the ways different consumers behave. And of course nowadays that's even more complex and getting that so even more quickly with cloud delivery of television content on demand, customized personal recommendations in the TV world that were never an issue before, never an opportunity before, I should say so rapidly increasing complexity with the same number of hours available at board meetings, the same number of senior executives, but nowadays spread thin across so many different routes to market and methods of engagement. Whereas in the old days it all concentrates on being an expert in just this one route to market. And the good news because opportunity or the bad news for complexity, for anyone who's 40 or older maybe, is that look at the youngsters and you realize this is going to change even more quickly going forward. This is only going to get worse or better, depending on your glass half full outlook. So you've got these digital natives, of course, who've grown up using technology and expecting technology. Brilliant. But what's more, older people like myself and some of us in the room. Not many of us in the room, I should say, but some of us older people are starting to adopt those behaviors of the youngsters. So that's accelerating the pace of change. You've got the youngsters growing up, of course, and more youngsters being born, so they're accelerating the pace of change. You've got proliferation of connected devices, maybe reach saturation in core markets in the US but around the rest of the world as well. And you've got this amazing access to data technology and code, cloud computing, open source data algorithms, APIs, which means it's even easier now to build new services and technologies than it ever was before. So they're going to come along a lot more rapidly. And then you've got this digital native feedback loop where the digital natives now are plenty old enough to build their own apps and experiences. No longer requiring us older people to build apps we think they might like they're building their own. So if you think the world's getting more complex and you think it might calm down, you're wrong. It's going to speed up, it's going to accelerate. So we all have to have strategies in place for how we understand this rapid change, how we think about what that means for our businesses and our products, our routes to market. We have to have plans in place in a crazy, fast paced, moving world. Let me give you one example of this in action at the BBC. Again, think about a corporate level or a brand level strategy. So for Planet Earth 2, one of our recent shows, well, who would have seen, there's a viral clip that went viral. You know what it is already? There's some snakes chasing some iguanas on a beach. Who saw that clip? Yeah. Okay, some of you are just too tired to put your hand up, I think, but that's okay. It's late. So, yeah, planet Earth 2, fantastic cutting edge natural history documentary. And for the show, we partnered with Snapchat to broadcast like mini episodes of the show, four minute episodes cut specially for Snapchat in lots of different ways, including fresh material, but kind of a four minute version of the show available at the time of linear broadcast of the show alongside it. And sure enough, millions of young people are suddenly engaging with natural history content on Snapchat every week. Absolutely fascinating. Who'd have thought just a small number of years ago for natural history content, that Snapchat would be an important part of how we go to market? I mean, we definitely wouldn't have done. And it took a lot of thinking, a lot of work and a Lot of analysis to really convince executives in a big old media company that this was an exciting opportunity, that the world is changing and how different people consume content is changing. Of course we do short form versions of it, and you saw the viral clips, of course we do Facebook and Twitter. But to get people to understand that alongside linear broadcast, which is our kind of magic window, alongside that window, we should cut special versions for a whole new platform with a whole new paradigm of what consumption means, what engagement means, what access means. A radical change in many ways to what it is to be a TV production company or a broadcaster. Really fresh thinking, just again, one simple example of how you need to think differently about these new technologies and the routes they might enable for your content to reach people in the world. So rapid innovation and a need to adapt at a corporate level, that makes sense. But tell me more about this complexity. And you know, I alluded to the fact that this complexity, at the same time as challenging us and making us think about lots of different routes to market instead of one or two, is providing immense opportunities. One of the challenges is that old business models, like the TV business model, aren't dying out, they're just changing. So the world would be okay if all these new routes came along. And if young people all watch natural history content on Snapchat, it would be a very simple place to be a TV business. But they don't, of course. Many of them watch on Snapchat, then go and watch the full show on linear tv. Or maybe they wait for the Netflix second window. The old channels don't die out, they just change. And we need to still support all of the old channels as well as all the new ones. And for example, in TV world, many people have been saying for years that linear TV will die. Linear TV is dying. The electronic program guide, the method of finding TV shows is going to go away. Turns out not to be true at all. Completely not true at all, actually. It's vibrant. There's tons of innovation that digital technologies allow around the program guide, and it's actually thriving as a way of finding content, reducing in size. Sure. And other channels are coming up alongside it. Absolutely. But this old world where we thought, oh yeah, the old channels will die, we can all focus our attention on the new channels. Absolutely not the case at all. We have to do all the old things and all the new things as well. Here you can see it in Nielsen numbers, looking at the hours per day spent consuming media, and you see that in the most recent data, thanks to the digital consumption of media there's actually a growth in, in how much media people consume. Yeah, sure, the old ones, TV is decreasing. Absolutely not decreasing to die, decreasing to change its role in people's lives as new channels come and grow the pie of consumption. So what this means, what all these fantastic new digital channels like Snapchat and other methods of consumption and discovery mean, is that even big old fashioned media companies like television companies or book publishers or record labels, the kind of businesses I spent the last few years working in, even those businesses are now swimming in the kind of data that we've been talking about earlier today. They're all using cloud computing to manage that on some level. They're all using some level of data science on that data. Brilliant. And in many cases there's great examples of that in action. Some of like what we saw today in every business you go to, I bet there's some great examples of how these technologies are really making a difference. In my experience, that often doesn't add up to where I said we'd go in this presentation, which is the kind of corporate level strategic thinking or the kind of brand level thinking. So sure enough, we're great at advertising optimizing at the ad level, optimizing at the message level maybe, or the channel level. In this world of complexity, how do you bring that data together? How do you tease out insight from it? And how do you make better decisions at the corporate or at the brand level? That's what I want to talk about. Of course, we're swimming in choice for how we do that. This is a fantastic diagram I look forward to every year. Even though you know exactly what it's going to show you, which is that there's loads of choice in the world. So this is the 5000 marketing technology companies available to us all today. So if you have a problem, you need marketing technology to help you. The answer lies on this page somewhere. It's brilliant. And I say every year because if you look back over time, the message is really clear. Every year there's a significant increase in complexity in the market. There's hundreds of new marketing technology solutions out there. There's one for every possible problem that you could have. So our problem in trying to navigate this world and trying to bring it all together, trying to tell stories that make a difference to our businesses, isn't the availability of technology. The first problem is how you bring that new technology into your business, how you work with all these vendors and how you think about what works for you, what makes sense. And my argument is that you need Structured ways to do that. You need structured ways to go out to the market and ask for ideas for new technologies that you should be embracing. You need structured ways to meet those technology companies, talk to them about your issues and opportunities. Beyond that, you need structured ways to help them to test their goods and services on your business, to try them out on business problems individually, see how they work, refine the offer. Often, because they're not fully baked, they're still working on their propositions. So refine what it is that they're doing and work out how they slot into your business alongside your other tools and alongside the business processes that you have. So how we do that at BBC Worldwide is through a program called Labs. And we do exactly that. Every year. We invite companies, not companies we've specifically invited, but in general we say, hey, who wants to come tell us how to run our business? It's quite fun. All these weird, wonderful, wacky companies come and say, I think you should do this. Let me show you why. And here's my tool that will help you. And sure enough, every year, this three or four or five that really shine, that come in and tell us something we hadn't thought about, we didn't know we needed, we try them out. We give them office space for six months. We give them a small budget to play with data, access to staff and meetings, a safe space to work, and some time to prove themselves. Every year, three or four companies really shine and become a really cool part of our business as a result. I'll show you a couple of them. So this is my first must. I can't imagine navigating the world with all of these companies, all these opportunities, and all these new challenges we have thanks to this increased complexity. I can't imagine navigating this world without a program, a safe space, a small budget, some time, and some desk space to give people that chance. Many of them don't work, many of them fail. Many of them come in with grand promises and end up leaving having delivered nothing. That's okay. Safe space, safe budgets, a chance to innovate. This has been a really cool part of how we've kept on top of this changing world and kept on top of the kind of technologies and tools and capabilities that are available to us to help us navigate it. So I want to show you what some of those look like now in terms of. And then I picked ones with lessons that are applicable generally. So specifically, they're interesting case studies, but the general lessons, I think are much more important for you. And my focus Here, the kind of tools I'm talking about, again, it's not ad level or message level optimization. It's tools that are going to help with strategy or brand strategy at the very least. And of course, much like the market overall, there's tons of innovation. So much like the TV example I gave earlier with the simple old world where we had just a television and a computer and that's all we had to really worry about. We're having the same message here, of course. So in the old world, in the tv, we used to all lean on Nielsen data every day. We'd run surveys alongside Nielsen data, but we basically had two data sources to bring together to tell stories. And those two data sources told us most of what we need to know about our TV shows. Of course, that doesn't work anymore. Of course we need more than that. So even if you just look at the brand tracker research that we do, we've always done, no longer is that anywhere near sufficient to warrant a big presentation. At the end of the brand tracker project, you have to bring other data from all these other data sources in in order to really understand what's going on. No single data source is good enough. So we've banned any presentations at the end of the brand tracker. You're not allowed to do that unless you bring other data sources in alongside to tell the story. Which means we've also tried to ban just looking at Nielsen data alone without bringing digital Data in and BrandTracker and all the other data. We've also banned the social media team from sending out their analytics reports in isolation of all the other data sources, because again, they just tell part of the story. They're an important part of the story. Of course, all of these are really important parts of the story, but you can't run a brand or you can't run a company looking at individual reports one by one. You have to bring them all together to tell the story. We've banned the CRM team from sending their analytics reports out, which they used to love doing. Again, a part of the story. And of course the web analytics and hundreds of other teams like, you can't run a brand or a company that way these days. I think you could. A few years ago. I certainly did. Worked pretty well, doesn't work anymore. It tells you such a small part of the story in this rapidly increasing complexity. You have to bring them all together. So instead of the brand tracker audiences debrief, get rid of the brand tracker. Let's think of it as a brand performance debrief. Let's bring all the different aspects of brand performance together, which is first party data, third party data, social data, all the other data sources. Let's not just worry about the audiences part. The insight team or the audiences team like to just talk about audiences, but there's something a bit bigger than that, right? Finances are kind of important and the insight team often don't get near the finances. They have to go, that's a separate meeting, that's a separate presentation. So bang the finances together. You have to look at the economics of the business, not just the audiences. It's another big challenge for lots of people. And then you have to bring it all together somehow. So here's my pattern for how I think about bringing it all together. You have to have your behavior data, your first party data, your social data, your financial data. You have to bring together research. This is another huge mistake, right? Just looking at first party data without looking at research. There's always things that aren't in your first party data. There's always attitudes and behaviors and reasons that aren't captured in your database. But if a first party behavioral data team don't have access to the research, if the research team's a separate team, you fall down a trap. So research and behavior together. Plus these days, with the complexity of our markets, with a number of data points and variables, issues and challenges, and brands that we're all trying to deal with, if you're not leveraging data science to automatically interpret at least some of your data to find patterns for you that you wouldn't have thought to look for, or to spot trends, or to tell you that trends you think are trends and not trends. If you're not using data science to leverage to look at your analytics, you're missing a trick. You're not scaling the way you need to scale or you're interpreting data in the wrong way. Again, I don't think that used to be the case. Just a few years ago, you could look at relatively simple cross tabs from data and draw quite powerful conclusions. But now there's so much data, there's so many variables and so many routes to market. I don't think anybody can be expected to do that at scale without data science. And it's so freely available with minimal skills, it's so freely available now, you're kind of crazy not to as well. And finally, the bit that you could often forget in a world of data science. You need people with great storytelling abilities. You need researchers at their core that are trained to find patterns in data and to bring that to life, to tease out the so what from the data and to help busy executives to see the conclusions and to make decisions as a result. So I argue, you need to bring all these things together in a meeting. If I look back at the projects I've done or I've seen in BEC worldwide that have been really exciting, that have really made a difference, it's when all these things come together, I'd say most of the time we miss one or two of these things out. Yeah, we do. This is best practice. It's hard to do this every day. Most of the time we miss one or two of these things out. And that's the meetings where you say, look, it's a really good presentation. There's some great data here. I just don't know if it tells me what I need. I just don't know if it's the right answer. I just don't know if I can take action on it. Great data, great story, great insight. I just don't know if I can take action. For me, it's usually because one or more of these things are missing. So I'd urge you to. Next time a report comes across your desk, next time you get invited to a meeting to look at some insight or some data, or next time you hold a meeting, I'd ask yourself if you're using all these, and if you're not, I'd ask yourself, am I missing a trick? Is there something missing? Yes, it's bloody hard work. It really is. Different skills, different people, and much more time consuming than the old way, where we just look at one data set and you make some conclusions. But it's necessary in this world, I really would argue so. A tough lesson, but a really important one. It's a pretty one as well, isn't it? If you believe this conclusion, though, if you believe that you do need to bring all those things together, it has some more profound implications. I mean, one is that gone are the days when you can lean on a full service agency with access to a certain set of data, let them go run a project, come back and tell you what, what you should do differently. Gone are those days. Unless you've given them access to all of your data across all of the verticals we just talked about, they're just giving you part of the solution. They aren't able to give you the so what? Think about all your channels and the interaction. If you believe that conclusion, this is a pretty profound. If you believe what I said in the last slide, this is a Pretty profound conclusion. Another is that colleagues with similar single areas of expertise are in some trouble. In this new world, if all you can do is social media analytics without thinking about research or the economics or the first party data that you have, if all you can do is that, then again you're relegated to telling just part of the story. In this new world, you need to either adapt the skills to branch out across other data sets and other skills, or you need to accept that your role is to give your data to somebody else and for them to bring it together with all the other data sets and to tell the story. So we've had people at BBC who are very good at one specific skill realize they have to now step back from presenting to the executive team from presenting the results. They have to step back and give their data to somebody else who brings it together and presents. That's a big change. That's a big change. We can't have just part of the story being told and we can't have decisions being made on just part of the story. So again, seems maybe simple on the previous slide, but some scary conclusions. I'd say the same. We talked about researchers, I'd say the same about data scientists who can easily and often get too big for their boots, thinking that they've got all the answers and that their work is the be all and end all. A fantastic and amazing piece of the puzzle, but a piece of the puzzle nonetheless. They have to team up with researchers, social media analytics, first party data analysts as well if we want to tell this rounded story. So really tough conclusions. And I think you'll see these being broken every single day in business, but when they're not broken, when they're followed, then you get much more powerful results. So to come back to the complexity, all these systems and routes to market and social media channels and blah blah, blah. One byproduct of this obviously is the vast amount of data that's being left behind by people in their everyday activities. Google searches, Facebook clicks, all these data streams, I think they're hugely exciting in the media industry and I think we're just starting to properly realize it. So until now they've each been used in the way I described earlier, which is to optimize individual activities. So the Facebook data is fantastic for optimizing Facebook. Twitter data is fantastic for optimizing Twitter. But how do you bring them all together? And I think the emergence of true data science at scale, at low cost or free, with open source algorithms, and the addition of the cloud computing and data science Working together means we're starting to see the power of this data and unlocked and offering genuine insight. I think it's going to change everything about how we think about brands and strategy. Like I say, in the media industries, I don't think most executives get that yet. Most don't see that yet. But there's enough signs. I'll show you. I think this is happening. I think it's real. I think we've been promised it for years and years and years. I think it's now coming. So particularly, I think what that means is we're going to see a switch from the old world, where the media industry, the TV industry in particular, lent so heavily on Nielsen data, plus a little bit of survey research, to a world where they switched to thinking about all this free social exhaust data that's out there in the world, but swirled together with data science, not just on its own, not just counting numbers and looking at trends, but swirled together with true data science. So one example of this in action in the TV world is a company called Parrot Analytics. Parrot, like the bird. Has anybody heard of them? A couple over here. Thumbs up or thumbs down? Yeah, 50. 50. All right, good. So, yeah, I think they're the future of television data, and I think thumbs up, thumbs down is about where most people are. It's kind of exciting. I don't think I understand it right. It's new. Here's what they do. They take all of the social digital exhaust data that's left by people looking for, looking at and engaging with TV shows. So when you search for it, when you look at the Wikipedia page, when you tweet about it, or you follow an account, or you watch a video on YouTube, all of this data, which again, has previously been individually useful in those channels, and they swirl it all together for a particular TV show and aggregate it back to say, how much demand is there? That's their phrase. How much demand is there for that TV show in that country on that day? So previously, when Netflix measured their own consumption, Amazon measured theirs. TV channels, through Nielsen, measure theirs. Suddenly, we now have this single metric for every TV show, whether it's on air or not, in every country, again, whether it's on air or not. And for every single day, again, whether it's on air or not. And we can say, how much demand was there for my show versus the other show. Doctor who's not even on air at the moment, I think, yet. Just finished. We had a big announcement yesterday, if anyone hasn't heard already, and I can look now. And say, well, how much demand did that generate for Doctor who? Even though it's not on air, it's not measured in Nielsen right now. Netflix have their own measurement. We'll get that in time. But I can say, how much demand was there yesterday relative to Game of Thrones or other shows that I heard might be on air? Maybe Game of Thrones, I don't know. I heard of it. So, yeah, digital exhaust data freely available has been for bloody years. Finally, with fantastic data science, it's being stirred together and it's forming this new metric which I think is going to fundamentally disrupt the old metrics of our industry. Really exciting times. And if you look for Planet Earth 2, ask a question, who is the audience? Again, we would have used Nielsen data, we would have used survey data on top of that to understand the audience. Now we can use this freely available data in new ways. Who's heard of Affinio or who uses Affinio? You're on it with all these, aren't you? Ok, great. So I wasn't looking over here, I was obsessed over here. Who else? Is anyone using Affinio? Okay, no. Oh, wow. All right. Anyone heard of them? You should. It's amazing. Not because you should necessarily subscribe to them, other services are available. But the general methodology here is fascinating and groundbreaking, and it's a lesson for all of us. What they do is to take, again, data that's been available bloody ages. This is people who follow the BBC Earth Twitter account. Just Twitter, follow a list. We've had those forever. They join them together so that each person, they know all the other accounts that they follow. And they use some really clever data science to cluster people according to the other accounts they follow into groupings of people with similar interests. And suddenly, really, instantly and really insightfully, you can work out who the different types of people are that follow your account. So for BBC Earth, there are a bunch of people. They're all very, very similar. This dark green lump here, you look into it a little bit and you see they're all animal lovers. That's the thing that unites them. That's who they are. Therefore, that's what they want from BBC Earth. If you look down at the bottom on the left, there's another green group. They're also very tightly packed together. They're very similar. They've got some links to some of the other groups, but they're different. They're very different. Click on them and you understand very quickly, they're all photographers, all sorts of different types of photography. That's what they are, it's what they want, that's what they follow and that's what they want from us. Different from the science and biology group, different from the sustainable and environmental group, different from the big general audiences out there. So again, super easy, super quick, free data, science on top. Lovely jubbly. You can really easily see who is my audience, who are the different groups out there. A whole new level of insight that we've never had before at this level of scale. Globalness, richness, actionability. Never had this before. Instantly available. Again, what's the lesson? 4 free data plus really smart data science on top of it. I think it's going to change everything about how we run our businesses. Again, from brand strategy, kind of corporate strategy point of view, two great examples, but there's many, many more. Then to bring it back to the topic I picked for this today. Okay, so there's great tools and technologies, there's great data out there. We need to keep on top of all these new technologies and startups. We need to bring them into our businesses. But how do we think about this day to day? My key recommendation is to democratize, which is hard for an insight team or a research team or analytics team to say we want to be the experts. We want to stand at the meeting and tell people what the brand strategy should be, tell people what the audience plan should be, tell people how a campaign is working or not. We want that level of importance, that level of say. But I'm saying we shouldn't do that. I'm saying we should instead empower everybody in the business to have access to these tools and capabilities. We should help them to understand their audiences and their business problems. Let me show you why. And if you rewind to 1997, Garry Kasparov is beaten by IBM's Deep Blue at chess. The moment where we might have drawn exactly the wrong conclusion about the role of data and technology in our lives, we might have drawn the conclusion. In fact, I think we probably did draw the conclusion at the time that computers were going to outsmart us. Recently, Google's DeepMind was it beat the world champion at go, which is infinitely more complex than chess. The games keep falling, AI keeps getting better and better. Computers keep getting smarter and smarter. Will they take over our decision making? Will they beat us at the core, things that we're good at? You might have drawn that conclusion back in 1997 from the chess game. But if you fast forward until recent times, a couple of years ago, so what? 15 years at least of chess Computers and people working and playing together and working out what the answer is, what the steady state is, what the equilibrium is, what the optimal solution is. You realize that's not the lesson at all. It's not the lesson that computers are going to take over and make decisions for us. There's two key conclusions from modern chess that I think apply to how we use data in businesses or how we should optimally use data in businesses. The first, which is kind of obvious, I think you wouldn't be here otherwise, but bear with me. The first is that you have to partner with data with computers if you want to win. So teams of humans plus machines dominate even the strongest computers. It's kind of interesting. Human strategic guidance combined with the tactical acuity of a computer, it's kind of interesting, was overwhelming in recent years. So it's not that computers be humans at all, actually. It's that the pairing of a computer and a human together is far stronger than even the best computer. Isn't that fascinating? The next conclusion, I think is even more important, and it goes to the heart of what I want to convey today, which is that it's not just about having a great chess computer. It's about being really, really good at how you use that computer. So I said, partner well with data. So a weak human plus a computer plus a better process was superior to a strong computer alone. Makes sense. But more remarkably superior to a strong human and a computer with an inferior process. So it's not just about. It's not about having the best chess computer to work with. Not that at all. It's not about having the best data analytics systems or the best data and analytics team. That's the parallel. I'm sure they've got to be minimally good, but then you've got to work out how you work with them. You've got to work out what your relationship is, what decisions you defer to the computer and what you don't, what decisions you make, how you lead them, what questions you ask, when you use them, how often you use them. You gotta get really, really good at partnering with the chess computer here or data analytics in our business. And this is a mistake I see made, like all the time, which is either leaning on data too much, forgoing that tactical acuity, or leaning on data not enough, or just leaning on it in a very kind of slow and unprocess driven way. So the title of the talk was this, it was augmented experts. And I think this is the future that we need to build if we want to Rock. In this world. There's a phrase that some people have used, you know, let's not let computers take over, let's keep human in the loop computing. I think that's the wrong phrase. It should be computer in the loop humanity. So people making decisions, people are in charge. We're making strategy decisions, we're making brand decisions, we're making all these decisions in businesses, but we've got computers in the loop, Computers helping us make those decisions, helping to optimize, helping to challenge us. Rarely are we very, very, very good at that. That comes back, of course, to when I talked about the BBC Worldwide Labs program, bringing these startups in, giving them a desk for six months. There's no good bringing a great startup in and not gelling. It takes at least six months at the BBC, I tell you, to really get good at how we use technology. New reports, new data, new insights. But you've got to invest that time and really get very, very, very good at it. If you want to make the best decisions you can. And people do that all the time, of course. So all of us, every day, reach into our pocket, help, ask computers, data, AI, machine learning, ask it to help us make decisions. But we're very used to making the decision and leaning on data to help us make that decision. It's the same at work, of course. Course. And when you get that right, when you invest the time it takes to get really, really, really good at using all these new technologies, given the time and the space, invest in processes, that's when you can make the better, faster, smarter, more global, bolder decisions. And that's pretty much how I see us working towards the BBC. So unprecedented innovation, more routes to market, more routes to measurement as well. Complexity in this understanding and the insight that's available to us, but a need to drive towards that insight and the way it's worked for us, the way it's worked best, the way it's had, the biggest decision is when I think we emulate the world of chess, democratize it, help people to get really, really, really good using the insight tools. And that's my big lesson for today. Thank you. It.