Media's shift from eyeballs to outcomes

[Music] Hello and welcome to this episode of The Warp Podcast, my name's Alex Bramsel and I'm the head of Content for Warp Media. Today we're going to be talking about a major transformation underway in how media is measured. Historically marketers have tracked whether their campaigns have been seen or heard, attempts to correlate that exposure with tangible actions work generally retrospective and in most cases fairly hard to prove. Even in digital media techniques like last click attribution have been discredited as only telling a small part of the story of how customer went from discovery and awareness to conversion. More recently however big tech platforms have used AI and machine learning to begin to enable brands to plan and buy ads against specific business results from app downloads to product purchases. This shift to outcomes has profound implications for the wider media industry. In just a moment I'll be joined by Samir Moda, the leader of the outcome measurement innovation team at UK Broadcaster ITV and Kate Brinkley head of digital planning at the specialist works agency. But first is a quick word from our sponsor. Good news, decades of benchmarking insights and best in class case studies are now available at the push of a button. Lions Intelligence is a new operating system for creative marketing excellence. With state-of-the-art AI technology and established marketing insights from the work, contagious IQ and Warp, all in a single adaptive workspace that evolves with your thinking. Save hours of manual work as share your thinking with clarity and confidence. Lions Intelligence is where groundwork becomes great work. Insights, strategy, creativity, execution, all in one place. To learn more check out intelligence.lions.co Hi Kate, hi Samir, welcome to the World Podcast. How are you? Hi Alex, love you to be here. Good this morning, thank you, how are you Alex? Very well and I'm very glad that you're here to help me explain this topic and explain what matters to our listeners. Should we dig straight into definitions, we love a definition at walk? What do we mean by outcomes in the context of media and advertising? Because it's one of those words that might mean different things to different people. Samir, do you want to kick us off? I think a very blunt definition might be something that a finance person is going to care about. Because that interface of marketing function to finance function is a critical one. And if it's something that the finance team would go, yeah, that was worth doing, then I think it's an outcome to agree Kate. Yeah, I guess to build on that slide is that people use the term outcomes in lots of different contexts. So you'll talk about it in terms of business targets, marketing objectives, you can use it in channel optimization, in pricing models. Like there's all sorts of uses of the word outcome. I agree with Samir that a very top line is about knowing that you're achieving something that means something to a business versus getting bogged down in a media metric. Yeah, I think there's an element of linguistic creep that's happening with it at the moment because two things I suppose. So if you think about the, I think it was the COI, if people remember that, that sort of popularized this framework of output, outtake outcome. And the output was the sort of the raw delivery, the outtake was the changes in people's brains, until availability, if you want to call it that. And then the outcomes were downstream of that things that this would care about. But in the sort of back and forth at the moment, it feels as though people are forgetting that framing and then saying, "Well, outcomes can be lots of things." And they can be very tactical things and brand, I brand measure is an outcome and everything's an outcome. And you say, "Well, no, we need to be slightly firm about this." And the lesson of the past 20 years is that when you show up with something for your medium that shows non-marketing people what it did, that turns out to be very compelling. You've hinted at that. Could we get a little bit of a potted history of where we've been and where we've come from with outcomes? What some of the platforms have done to sort of raise the agenda around this? And what why there is now an outcomes team at ITB, for instance? What are the ways of thinking about this is, I remember from my time and then the Google sales org, which is a fairly improbable place for me to be, but was hearing some of the training around how to sell such. And for years it was just this beautiful narrative because you'd sidle up to someone and you go, "Oh, nice website, you got there, I do, would you like to buy some, buy some clicks for it?" You know, we can first click the cheapest you'll ever buy and you build from that to there's the whole narrative around how searchers are saying, "Oh, look at what's happening now and are you spending as much as you should be because really, you should, how much is a website visit worth you? Oh, well, if it's worth that much then you should be thinking about spending more per click and kind of spending up to the diminishing returns point and then then they'd come back with another measurement product bolted on top and say, "Oh, wow, we think actually we're sending people to your physical stores and we've got some evidence of that, so let's add that, oh, oh, look, you need to, you can afford to spend up to a higher level of diminishing returns now, so we'll just unlock that budget shall we?" And I'm presenting it slightly solitally, but that was a revolution for vast, vast numbers of advertisers who previously had had nothing like that. And you know, we can get into, was all of its stuff that was being contributed by search, you know, the pizza parable and all that kind of stuff, but broadly speaking, it was a a very compelling sale in an environment where there'd be nothing like that. And I think kind of the difference now when we look at that progression is I think it it may be changed the perception of the standard of measurement, whether perception is always reality, I think is a slightly different conversation, but it changed the perceptions of standards of measurement, but on the downside it also isolated measurements, so it started to put things into very specific boxes, with kind of disconnected visibility across various platforms that would make similar claims to what Google would be able to show in terms of, oh, we're delivering this in terms of sales, metal would come back and say, yes, we're also delivering this in terms of sales in a different platform measurement. And I think the likes of Amazon are going that way now in terms of what they can see and there's there is a huge amount of power in that, but what it meant was that everything started to be seen in silo. And I think when you shift back to an actual business outcome, when you're looking more holistically at what you're delivering, it gives you that opportunity to evaluate what you're delivering against an outcome because nothing works in isolation. And as marketers and as agencies, like a lot of what we spend our time working about is what's contributing to delivering a business outcome. And when you get tied down into media metrics, you just start seeing things in isolation and your biggest unlock on gross is understanding the joins on how those things work together. And I think that's what coming back to a business outcome enables. It slightly takes away from individual attribution and individual proof points that are all disconnected to what actually gets your business working in a whole. And to be honest, quite often that is something that isn't a direct media deliverable or a channel deliverable, do you know, pricing competitors, the economy, things move outside of media. And I think that's the nice conversation and focus that comes back when you're thinking about business outcomes is you stop getting bogged down in channel metrics and start looking at what's working on why and how you then use that to really drive business growth rather than worrying about whether if your CPC goes up to pound, it's actually going to deliver you more sales or get you past that point of diminishing returns and starts delivering negative ROI based on what a platform is saying. Because it's interesting, you say that Kate that obviously Marx has got to be confident in the inputs and that we've got to be careful about the extent to which maybe that we hand over the job to AI. But all the evidence from an investment standpoint suggests that brands that marketers are all in on this. That we're seeing huge upswell of investment with the likes of Alphabet, with the likes of Matter and Amazon. We've seen it in their latest earnings calls that marketers trust it. They do believe what they're being told. Sam, do you think that is the case, is that your perspective as well or are you seeing something different depending on the type of client? Yeah, I think it is an interesting one. On the two bits, so on the machine learning side, there's an interesting evolution that's going on. On the client trust side, you think maybe the kind of the working mental model that a typical book subscriber has is off the world where there are people in Kate's team who are capable of doing a brilliant job for advertisers. But that's the minority advertiser experience now. There was a period in Metters reports when they would add a million new advertisers per quarter, whatever it was. And so the modal or the median experience of an advertiser is probably only using one or two platforms, probably doing it themselves, probably doing it as an adjunct to other things. And pretty trusting of the stuff that they see. And it is very much this, okay, I haven't got much time. I've got to set this thing out. I'll chuck some money in here. I'll chuck some pictures from my website in here. Tell it what I care about. And it's guided me to that quite nicely and off I go. So yeah, in terms of the weight of spend, most of it is not spent with or by people who are giving it a critical eye. But they are all people who are finding that it is working for them. I used to work with a chat at a wine chair who was a sort of search technician by trade, but also had a side business selling water bottles. I won't mention it in case he doesn't want to be mentioned, but it's now a very successful and big water bottle business in the UK and beyond. And I said to him, so this is interesting. You do your own search stuff and then you do it for your day job. What's the difference? He said, well, when I'm doing it for work, there's lots of people crawling all over every decision. And it's like, am I going to add this word or that word? This is 10 years ago, I'm like, so it's obviously coming up a bit since then. And then we'll argue about that and we proceed very slowly. When I'm doing it for myself, I just flick all the machine learning to on because it turns out it's quite good and I'm making lots of money, which is why I move out to quit. And sure enough, that's what happened. So much as I, with my ITVHON, might have an ambivalent relationship with them, they are some of the kind of glories of modern data science. And as Kate says, that's not a new thing. For 15, 20 years, they have been taking the best of modern scale machine learning and applying it to build planet scale models that buried inside them no more than any or possibly all humans have ever known about the relationship between content and context and audience. One of the things I did a bit when I was a Google was trying persuade people to let us use the production models to expose what they had learned about people that were surprising, which is a great fun thing to do. But coming back to this is the technical bit. There is a technical gear shift happening. And in fact, season Lee, who was on the the investor call for meta, this just the one just gone, said something that Matt Steiner, who leads all that stuff for them, had previously trailed end of last year, which is they are taking, I don't know if you could call it the piece dividend. It's probably the war dividend of large language bottles and applying them to rebuild their ad stacks because two things have happened. Firstly, all the generative creative stuff has generated, has created so many different ad variants that it's clogging up their own pipes. So when you go from choosing between a million ads that you and campaigns that you might be serving to a billion, the current ad stacks are creaking because they are 20 years old. And generally you don't fuck about with something that's making you billion sub-hounds. But now they are doing exactly that. And they're saying, right, we're going to take the sequence model architecture and apply it to our ad products. And Matt in particular said, well, there's no new data. Of course, there's no new data because they've got at least five years worth of outcome data from millions of advertisers and behavioral data down to a very low event level for billions of consumers. And in the same way that when you train a language model, you say the cat sat on there. And then you say, and the model says dog, you say, no, bad model. The cat sat on the mat. Yes, good model. Do that a zillion times. You've got a large language model. Similarly, give it a sequence of scrolls and events in Instagram or whatever. And it will start to work out some really spooky stuff. And that is exactly what they're doing. So I do think in my ITV work, I feel a massive sense of urgency, not quite verging on panic, but occasionally, because those products are about to get 10 times better. If we think about the shift even from chat GPT-3 to 5 or Opus 4.6 that landed last night, which is what I haven't slept, just think of that scale of speed of improvement that applied to ad products that are already really bloody good. And that is what we are facing as good old fashioned telly. We will get on to telling in just a second, but okay, I wanted to kind of come back to you and from your perspective, because, you know, Mark Zuckerberg's made absolutely clear that he believes the future of advertising will be as an advertiser. You hand over the keys, the money, and give them a little bit of a North Star, and they will do the lot and measure the lot. Are we moving towards that? Do you think? I think there's a difference between the power of some of these ad tech platforms to deliver great results from the media by, because to Smith's point, they are very, very good at targeting. They've got really powerful optimization models. They're good at understanding nuances. I do think that there is a difference between what the power of each of those ad tech stacks to do that in their own right and how you actually need to impact the way people think. I think it's very easy to get sucked into all of the benefits of the ad tech platforms, which are exceptionally powerful, and forget that we're talking to people. And ultimately, whenever we're going out with any level of advertising campaign in any shape or form, we're doing it to influence people in some way, shape or form. And whilst each of those individual platforms pick up on people-based signals as part of what they're doing, they still are only seeing themselves in their individual ecosystem. They're not seeing absolutely everything. And I think that's where you get a shift between really powerful ad tech that works within its own environment to how the docs join across a much bigger space, which is the reality for a consumer. And I think that's where there will always be a bit of tension between the power of ad tech and what we actually need to deliver in terms of shifting the way people think, their preference, their choices, when they come to you, it can't all happen in an ad stack because humans don't only exist within an ad stack, ad tech stack, sorry. And I think that's where there will always be a tension in a rub is that you can't see everything within a single ad tech stack at this point in time. Like whether that changes with developments in the future is maybe a slightly different conversation, but right now we're still talking to people and our tech is not the only part of that conversation. I did though though, it is a lot of it, it's funny, I sort of said I've got a platform or apologize here, but the vast amounts of time, I'm not making any political point, I'll leave that for the grown ups' ITV to do, but the vast amounts of time that we all spend on these things. If all you have about me is my behaviour on meta-platforms, do you know quite a lot about me? Bloody hell, yes you do. Yep, absolutely agree. And so yes, there is that silo over and yeah, and for sophisticated advertisers who can have a clever agency to help them, there is that, but the most common experience for most is one platform, maybe two, and mostly it just works and they don't need anything else. And so I suppose I worry about, and look, I started life, believe it or not, as a creative planner, writing briefs for teams in an ad agency, Jowel to Thompson, Barclay Square, you know, I have sat on the floor of a creative office amongst the beer stains, you know, arguing about the position of a comma or whatever, and taking poly boards up to a focus group facility in leads or, you know, I have a huge affection for that side of our industry, that kind of thoughtful side that is able to achieve, you know, cultural change, able to achieve all sorts of wonderful things through exact newscakes as thinking about people and reasoning about people and making creative leaps from where people are today to where they might be tomorrow. And yet, and yet it feels like at the moment in the debate, there's this sort of, in the back of the fourth, there's this sort of alliance of people who are formed of traditional media measurements, reach frequency, eyeball type stuff, people who kind of believe in brands, if you like, and the ability of brands to do things. And that sort of has become this sort of force of friction in a way, because all of those things get tangled up together. Oh, you've got to think about the long-term time horizon. Oh, you've got to think about brand, you know, but if you look at the language people use, it's almost like they're peddling a religion. I mean, I'll hear people talking about advertisers who believe in brand. It's not a fucking religion. Sorry, it shouldn't have to be that. And I think we've just been a bit preachy, and we're sort of just telling, I mean, much as I like, you know, I was at Peterfield and Les Bannett's first, one of their first things when they first had walked in the air of accountability. And I remember mind blown, amazing. Okay. But now we're still on stage, you know, Peter and Les separately now. It's like Simon and Garfunkel's splitting up. You know, they they will still talk about the very long term and how it's insane that advertisers aren't worrying more about that and so on. And Neil Bournemann, I think, got a spiky riposte of that at an event where he said, I won't try to do his accent, but, you know, he basically said, we need to measure better, not lecture harder. And it's just that the, I think it's worth at the moment, leaning quite hard the other way, just because there's so much inertia. And there are so many people who've got to cut their teeth on IPA papers and all that lovely stuff. As it I, you know, I've been a judge and stuff like that. I love all of that, but it's currently not helping us because we're not engaging in the other stuff. And you know, there's very few people like Kate who have spent quality hours in front of a meta or a Google dashboard or buying front end. I'm always appalled actually, sorry, I wish shut up, but I just, I do, I do this at conferences now and I'll say to people, you know, hands up if you've seen this screen before and it's the Google ads buying screen where it says, you know, welcome to a little shop of outcomes. What would you like to buy today? Would you like to buy store visits? Would you like to buy online sales? And the number of people that don't put their hand up when I say, have you seen this before? I just think that's appalling. You know, how could you not look at the screens where billions and billions of pounds are spent and and and still claim to be an informed marketer? Sorry. You know, I do think that's challenging. I also think the story that comes with that in some ways amplifies. And so, you know, there it's exceptionally accessible. Like there it's it's really, really accessible. And I think if you look to like, there's some very, very spot businesses. If you if you look to look to Google, they started with investing immediately in a measurement product in GA that everyone could put on their websites and that actually did a did a very, very good job in building people's trust in them as a business. And that also happened to showcase the performance of their channels very well. I don't really feel that a business as far as as Google did that by accident. And like, but and I think what they really tapped into there was they really understood the challenge to showcase impact of channel spend on business on metrics businesses care about. Like that is the pressure overall we're having to answer as a business all the time and they got there very, very quickly. And I think if you come back to that principle, it's really interesting. Like often you don't actually buy the CPA. You're not actually trading your media on a CPA for sales. You're still buying on a CPC. You're just optimizing to the outcome. That's a nuance, I think, that often gets lost in some of those conversations because the story is so strong. And yes, you can see within that same screen, this is how much money I put in. This is how many clicks add to baskets and sales I drove in there for my CPA is X. Like you haven't actually bought the CPA, you bought the CPC, but it's all in it's all in one place. And the fact that GA would then make that visible, the kind of website understanding tool that wider business stakeholders can see and understand is it's really smart. Like that was a really bright move in terms of connecting spend to results. And I think that is just the underlying challenge that's facing the rest of the industry. How do you understand that pressure and answer that pressure to connect spend to something that's understandable to stakeholders and that just circles background to this business outcome conversation in that we have to keep our eyes up on a business outcome versus getting drawn into those kind of really niche specific channel metrics. Just to speak on behalf of Walk for a moment, which is not often that I attempt to do so, but Walk as a platform has four plus decades worth of evidence and research on advertising, which shows that there is a better way of doing things that if you kind of move in that direction and you would hear to certain, it hears probably the wrong wrong word, but you follow certain best practices you're going to do better as a business. Ultimately in lots of large companies have been built on those principles. So we know they work. And that's to suage any marketing effectiveness experts that have been throwing tomatoes at their Alexa device in the last five minutes. So is this purely about convincing the people internally within an advertiser who couldn't care less about the metrics that we've often talked about, things I reach or whatever it might be, that it's not necessarily about doing worse advertising, it's about talking to them in a different language, or is it about doing things differently? I really don't think we've tried that strategy for 20 years, and okay, we don't have the counterfactual to compare against, but it doesn't really look like it's worked. It's very hard to draw any other conclusion that other than that approach of to, again, to Neil Bournemont, preaching harder, it just hasn't worked and it doesn't matter. It may be true, but if it's true and nobody's listening, it's irrelevant. And so the work that we're trying to do is to get to get telly to show up in places where it hasn't showed up, it hasn't shown up in the weekly meetings, it hasn't shown up in the courtly reviews. Yeah, okay, it's just just about shuffles onto the stage by the time you get to the MMD brief, which point you do, something in Kate's team is going, yeah, but I need to plan the next bloody campaign. Come on, hurry up. If you're listening to this podcast, you'll know that walk is the home of leading effectiveness thinking. And now, we're teaching you everything you need to know about the fundamentals of strategy. Walk strategy fundamentals is Walk's new digital course for junior or entry-level strategists. You'll learn from award-winning case studies and hear directly from top strategists and Walk's own experts, giving you the latest industry insights alongside proven, practical frameworks. Shapping up your skills to excel in 2026 with Walk Strategy fundamentals, available now at walk.com/learning. It's so weird just to take half a step back actually. For any listeners, particularly those, I guess, that are not based in the UK that might not have some context around what what ITV is doing in this space. Could you just give a quick, potted history of your role and what your team is doing and I guess what the aim is? Yeah, sure. So joined about five years ago and obviously you could see early signs of what's now more obvious in terms of the money shifts that we're carrying on. And we started with an analysis, more or less like, we were discussing, so look, it's not right to blame advertisers for what they're doing. They are doing the best that they can with the information that they have. So what if, historically, that the arrangement has been, we, the broadcasters, will pay for the eyeball measurement, will give you that, will have a jick so that it's done fairly and we can treat it as a currency and all those lovely things that happen in the 70s and 80s and so on. And whether those eyeballs do anything for your business is left as an exercise for the reader, you go figure it out, you go find a modeling provider or whatever, that's fine. And then, you know, along came platforms with this very different way of doing things we've been talking about. And so we said to ourselves, well, how can we get telly to show up in those places? So, you know, outcome, or all the words carry weight. So outcome measurement innovation team. And there's three jobs to do really. Firstly, you've got to make the stuff because the product to measure TV impact at scale in modern ways don't really exist. So we have to kind of invest in that. And we've had a pretty good run. And really, it's not ITV that we care about. It's the broadcasters as a total set, which is why some of the stuff like lantern that I'm sure we get onto is very much cross broadcaster because it's the medium that we care about, you know, scraping a couple of shackles off channels for or whatever isn't going to make any difference to my life. But defending the medium absolutely will. And then we, the second thing you need to do is help the sales orgs and help the industry understand those new bits of kit once you've got them. And socialize those, get them into ways of working and so on. I'm still, I know, I had a call yesterday with a lovely advertiser. Didn't know we existed as a team. What's the catch? And there's no catch. We will do good quality measurement work for you in conjunction with your agency for free. Oh, okay, that's really good. Yes, yes it is. And so we have the challenge of scaling that. And the the sales org ITVF say has been amazing embracing that because that can be a challenge. And then you have to do things like this. So the third pillar activity is try and share a lot of that stuff, make it available to other people will come up to talk about the price elasticity stuff which I think is a fascinating area. Because if in the the strategy I suppose is that we have to measure our way out of this conundrum. And if we can get more people in the industry using better techniques that surface the business impact of TV, then at least it's it's fair. And then after that, you know, let the advertisers decide. I think that that point about better measurement really resonates with me as well. Like ultimately my job as a planner is to make the best choices for our clients to deliver on a on a bigger business outcome. And the hardest bit to unpick in that is six different measurement silos that look at things in slightly different ways. And that's a that's a headache that there's a lot of people, lots smarter than me are trying to solve. But I think measurement is the key to this. And understanding in a more consistent way. And there's lots of lots of really powerful data out there. But it is all quite channel specific. And the silos are a challenge and they're not going to go away. There are there are different there are different ecosystems that exist and they're not all going to disappear overnight. So we have to operate around those silos. And actually I think there's the realities of measurement and what it tells you when is really really important. And the digital platforms have put this like a kind of lean on instant real time results that are immediately available. And but actually not every channel or platform operates in the same way. And definitely not every media operates in the same way. And I think setting up a measurement structure that's probably built out of several pillars of measurement because one solution is never going to give you a full answer. And understanding what it can tell you when and what its flaws are is is the really important thing in terms of a getting a view on what's going on. And be being able to explain that beyond a marketing team. And I think if we if we walk into a room and say measurements perfect. We're honestly lying. Like there's no way measurement could ever be perfect. And I think it's being honest about those flaws is is as important as pushing the results because then you interpret that with the right lens to try and make those best decisions. And I think I fully agree with some of you that measurement is is our way to finding the right answers. And but it's it's definitely not an easy easy one to solve. And it's where it's where you really need to acknowledge poor poor old MMM. Yeah, it's getting dragged in both directions on the strategic side. It's being asked to answer you know really big meaty questions about you know how am I going to get my growth in the future. And then at the other end you've got people saying yeah but I want I want a tactical answer by next Tuesday because I'm so used to seeing things live in the platforms and surely you can give me that. And it's it's it's this weird thing where you know and I'd argue that possibly more recent IPA papers haven't helped with this but it's it's it's become such a fixation for us that we kind of think oh yeah this is the gold standard. It's not. There's so much more under the econometric sun. But you know straightforward regressions you know it is in the old phrase it's correlation not causation. There's there's so much that is that is limited about it. It doesn't look at pricing. It doesn't look at competitor activity. And yet it's this sort of overburdened workhorse that we're trying to get to do to do everything for us. And so the innovation bit here I think is is really important which is to try and encourage development in areas that moves us moves us beyond that. Project lantern is an example of that innovation. Yeah so it's the thing about lantern is it's very deliberately a hybrid methodology because you do you talk to anybody who's worked in tele measurement that'll say yeah but if you measure in the really short term you're fucked because it's going to look terrible particularly compared to all that lovely platform stuff. It's like yes that's true equally if you wait a year or more to measure the effect then you've lost because you're just not in the conversation. So I'm trying and failing to try to get people to talk about the near-term impact. So not the next five minutes or the next five days but certainly within the week or the month there may be the quarter to push. And so the lovely thing about lantern as a geometry is it's got a tagged piece which gives you large-scale short-term lift in kind of platform style if you like but it is lift it's not just attribution it's an actual lift that you're getting and then the panel adds two things so the panel adds longevity because you can measure the effect out not just to the the seven days that you can just about get to with IP matching before it starts to fall over because somebody's cat sat on their router and it rebooted and the IP address got reassigned to somebody else but it goes out to the weeks and the months. And then the second thing that the panel does is it has loads of other outcomes. So you know it's right at the top of the podcast we talked about that and how there is a range of things that matter to businesses and it will surface all of those. So again you get this boost because you're you're measuring more in than near-term rather than letting the long-term do all the work for you because that's just a just I think that's that's an option for us today. We have to bring bring things back from the long term into the near-term and then we have to stack up the layers of value that we're getting to a point where we can then say okay look this is the total contribution I think you sure that isn't something you want to pay for and maybe hopefully pay a bit a bit more for the other one in that area is which is the most comically ambitious of the things that we're doing is I did I did a thing where we looked at the the number of IPA papers that mention price elasticity and it's shockingly low like three four percent in the past decade. So although we talk a great game oh you know adding a brand will increase your price firmness and reduce your need to discount and all those things we've done a rubbish job of measuring it and so one of the things that we're investing in is the application of a differently conometric technique called BLP which has properly grown up causal analysis. You have to feed it all the data for the category going back five years but then it figures out okay what was the contribution to the margin of these different businesses and that's the thing that where we're sort of starting to share a bit more now this year and to kind of give to people and say look we've built this model it's really powerful but it needs a bit of work to figure out what you can do with it would you have a go when you look at that for your category and see what you could do with it but in the end it becomes this kind of strategic simulation tool where you can say okay I now understand what the different channels were contributing to my margin all my CM3 or whatever now let me run some simulations to see okay what should I be doing over the next quarter or the next half year what happens if I do price changes and it simulates all the changes of the competitors and so on and that's one of the things that I'm I mean it's a slow burn but it's something I'm really excited about you know if I can come back in five years and say that we've managed to shift the industries addiction to a mm then I'll be quite pleased. I think that's the type of thing that's exciting as well because you're leaping you're leaping back to a business outcome and it's something that you're not waiting a year for every time I think one of the the biggest frustrations you have as a planner is that like you're having to write next year's plan before the mm and so on so like that you're all you're all you're too late to the party to make good choices and I think the that kind of speed of results is really interesting as an innovation point and I think that's the exciting bit right now right there that's what AI is enabling is there's a there's a chunk of innovation in the measurement space that is possible and can be accelerated with AI that's that's where we can get some really great learnings that help us use measurement to to prove the value of what we're doing or prove that it's not working and stop it either way is is is fine in my eyes like as long as we've got an answer that that people can can buy into and we can start to understand contribution because that's how we start start making better decisions about what we do going forward and I think where we need to get to is is and what I'm quite passionate about for us as an industry is that we get to a kind of commons not databases because that's the wrong thing what we get to is is large shared models that pool learning across hundreds of brands but that are not pay as you go and so you know we've got a tool that that surfaces one of these it's got 600 brands worth of data going back three four years all on a consistent basis and that is something that we're offering to people to use as an outcome planning tool so okay plan against lift but you can have access to this and use it yourself and and that's sort of that's the scale of innovation that we that we need and it just feels like we're getting dragged back into all these pole clutching debates about you know panel versus versus device on the audience measurement side when actually we need a concerted effort to to have shared learning and the bit that we do need advertisers to understand within that is yes okay the learning may not have come from your brand but it's come from a hundred like yours and whilst you're lovely or not that special and different and and you can benefit from that and that kind of that commons is the thing that I really hope we can pull off because the alternative is you've got to keep paying your interviewee to the world gardens I honestly could keep talking about this all day it's genuinely so interesting but I'm going to bring in one final question now and allow me to clutch some pals some walk related pals and we believe in long term brand building it's something we talked about an awful lot and that the brand long term brand building can help the performance end of the final as well outcomes to me strikes me as quite sort of short term ist potentially you know short term view and depend on category and and and and the campaign but are we moving to to an out without comes to a more short term this world where long term doesn't matter for marketing teams I don't think we are because I think it it depends on how you set those outcomes and and how you look at them and so there has to be value in brand and if you connect that back to the bigger picture and look at what brand drives beyond just pushing through users through the funnel you'll see lots of conversations about what it can add to bottom lines to preference and trust which I think's massively important particularly given the development of LLMS and how consumer behaviors change in the context to discovery like brands are a really important play in their in terms of in terms of trust and understanding with users price elasticity some is already spoken to and and there's a ton of research that shows that that brands are growth driver and actually you guys and covered a big piece by Cantar and a year or so ago and and I think what we have to do is note that all of those things actually contribute back to a business outcome but to make it really valuable we have to be able to join the dots between shifting a brand outcome and impacting a business one if we can't join that dot it's really hard to prove the value and I think that's where where measurement comes back into into play and also the brand and performance conversation is really important and there is more than one thing you're trying to do as a business at any one time understanding the relationship between brand and performance which are which are different things but both can ladder back to delivering your business outcome that's where that's where you can unlock an acceleration on growth from brand using it alongside your performance activity that's helping with the short term outcomes. It's a funny one this I think I think where we're heading is Little Bears Porridge so you know the the platforms are too short term but they're getting longer thanks to paramachine learning you know we've historically been maybe a bit complacent about the long term effect and not invested so much effort in measuring more more near term effect so I think there's that there's there's an element to which the measurement community with with in traditional media has been so content with the long run effect that it hasn't kind of come in a bit a bit near a near a term and then I think the other thing is it's actually about experimentation weirdly in that brand is is overloaded as a term and you know it often is a proxy for control and let's not let's not experiment let's make sure that we are consistent in all our different surfaces and you know that we're joined up because we know that having a single perception you know that that classic ad ad led converge on a single thought and then blow it out across lots of things and and get everybody thinking roughly the same thing about cheese or whatever it was and that's the bit that I think we need to let go of amongst the other poles that also need to be released but because it's it's okay if we're a little bit inconsistent it's okay if we're seen to experiment in public rather than me with my poly boards and leads because that is where you experiment now and that is how you get learning and I think the sort of the winds will go to those those advertisers those brands that are willing to do that a little bit more more openly rather than thinking all of that stuff has to be internal and then once we've decided what it is then we'll tell you the consumers what it's going to be and I quite like that I'm really sorry to say that's what we have time for today my thanks to cake and to Samu here remember you can subscribe to the walk podcast on your favourite podcasting platform and if you really liked it go on and leave us a review until next time thanks for listening