Marketing's Credibility Problem: What Attribution Gets Wrong and Incrementality Gets Right

James Lawrence: Welcome back to the Smarter Marketer podcast. I'm here today with Paul Sinkinson from Analytic Partners. Paul, welcome back to the pod.

Paul Sinkinson: Thank you very much.

James Lawrence: We were just talking very briefly off air, and it's almost been two years to the day that Paul was last on the pod, and the episode we had was a great chat, and it was one of the more listened to episodes that we've had.

So we thought it'd be awesome to have you back on and just, chat around what has changed, what hasn't changed in the world of marketing measurement in the last couple of years.

Paul Sinkinson: Yeah. It's always nice to be asked

James Lawrence: back. That's right. Must have done something right.

Paul Sinkinson: Yeah, exactly.

James Lawrence: I'll just do a quick introduction for those listeners that might not have heard the first episode. So Paul is managing director for Australia and Asia at Analytic Partners, a global leader in marketing analytics and commercial decision-making. Paul spent , just a little bit over the past decade running the business in APAC and helping some of the world's largest brands understand what actually drives growth and moving beyond surface level metrics to connect marketing activity with real business outcomes.

Obviously digital measurement had been becoming more difficult for a long time, but around the time that we were last chatting, there was all this talk of attribution breaking, cookieless world, signal loss the importance of creative consistency, probably a feeling that there was this kind of catching up of an understanding of the importance of brand.

But I guess fast-forward to now, what has changed the most in the last couple of years around marketing measurement from your perspective?

I

Paul Sinkinson: think people are realizing, you need to measure more holistically. I think there's, definitely a lot more say adoption of market mix.

It's interesting, right? Market mix modeling used to be one big amorphous segment. I think it's definitely broken into two different segments now. You've got- More reporting based, saying, "Hey, what did our marketing do for us last month?" Which is, like important because it shows the CFO that, you're doing your due diligence.

It, fulfills that board reporting part that's there. And then there's that forward-looking part where, you still maybe have people involved a little bit more, which is about value generation. So instead of what did our marketing do for us, it's what can our marketing do for us?

So I think, you're seeing the evolution of the industry. You're seeing it start to serve different needs for different people rather than just be one thing for everything.

James Lawrence: Yeah, interesting. When we last spoke, and I don't want this to focus exclusively on digital, but a stat that you said at the time was that 59% of online events are now missing due to privacy ad blockers and tracking changes.

Has that gotten better or gotten worse in the last couple of years?

Paul Sinkinson: It's gotten worse. But I don't know if that's necessarily the worst part of it because what's really ramped up at the same time is each publisher or each walled platform launching their own measurement system to go along with it, and none of those systems work well with another platform's system.

So it's almost trying to like one, you're missing data to pull into holistic measurement, but now marketers are getting more and more different measurement given to them. So you've got 15 different measures of the same thing which I think is making it like, incrementally harder.

James Lawrence: Yeah. It's so interesting, isn't it? Like 15 years back it's this idea that we've got so much data now, we can make all these decisions based on data, and the further that kind of evolution went, it's kinda hang on, we're now drowning in a sea of data and how do we actually, and from like a Rocket perspective, do a lot of work in GA4.

I think marketers have been trained to take as almost Bible the numbers and data that come from there, and then you're looking at platforms, TikTok, Meta, whatever it might be, all saying different things. Is it generally a nefarious play by whoever the publisher or platform is to overstate the importance of the work they're doing, or is it just genuinely a challenge or a difference in kind of perspective as to what numbers you put forward?

Paul Sinkinson: I think it's more nefarious when they won't share the data and don't want any measurement on it at all. So I think everybody who's saying, "Look, this is the way to measure it," is doing it from, you always do things from your own perspective, right? Because you live in that world, you think that's the way, that's the way that you should be planning the marketing.

That's the way that you should be doing it, so that's how you should evaluate it, and they never think outside of that world. It can be ... i've chatted with some publishers who've gone "Oh yeah, we keep trying to tell our sales team to, to think long term." And, In our business, that means people think it's 30 days after the ad was served.

That's like really long term. And it's because, like in their world, that's as much as you can measure on the platform with that data signal rather than using other techniques. There's nothing like nefarious about it, but it's just a demonstration of you think in the world that you're in, right?

When you've got a hammer every problem looks like a nail, right? And so I think that sometimes the help of more holistic measurement is that it gets even the publishers thinking outside of it because, to be honest a lot of them, they're missing some of the value that they're bringing clients.

So it's not nefarious at all. They're actually out there selling themselves.

James Lawrence: . So what are your best customers doing? What's their approach to trying to, wade through this sea of data, right? And try to make sense of it and to bring it together and I don't know if like making it homogenous is the, probably the right approach, but what is the way that your best customers or the best marketers that you're seeing, the approaches they're taking?

Paul Sinkinson: There's still not one thing. As much as somebody who sells mixed models is gonna say, "Oh, use a mixed model, it fixes everything for you." It's still, I think the best marketers are thinking about what it is that they're trying to measure, what the strategy is, and then thinking, "Okay, cool, here's the measurement for it."

Because mix is brilliant for that holistic measurement. It's the thing that's gonna tell you how much should you be spending on each channel, right? It's gonna do that better than anything else. The bit that it's not gonna do is, everybody tells like real-time mix models, pull the other one.

It's not gonna do real time for you. If you were trying to use a mix model to optimize your search terms, good luck.

It might be a weekly model, it might even be a daily model, fantastic, but it needs to be genuinely real time. 11:29 and 13 seconds versus 11:29 and 36 seconds.

What should you be doing? How should you be adjusting? That's real time. And a mixed model can never get there, so you still need that on platform measurement for that sort of a question of, within channel, how should we be optimizing as things are happening versus how much should we be spending on that channel overall?

Different question, different measurement. So I think it's still being led by that of what decision are we trying to do and then what measurement and thus data do we use for that decision?

James Lawrence: Yeah. It's definitely from our side, like there's no one right where this is the approach that works for every business and it's how do you put it together based on the needs.

In terms of like attribution, where like what are your observations there at the moment? When we last spoke we were looking at the challenges of attribution and is attribution dead? Last click attribution, at that time very much dead.

Is it still useful directionally? What's your kind of observations there?

Paul Sinkinson: Yeah, for the right questions it definitely is. If you've got that model calibrated, would I use last click attribution to tell me how much money I should be putting into search? No. No way in hell.

It's incredibly dangerous. Would I be using it once I've got that budget set, would I be using it to look at optimizing my keyword searches in real time? Yes.

Because it's the only tool that can really do it for you.

James Lawrence: Yeah.

Paul Sinkinson: So I think for the right questions it's still doing it. And I think, it kinda comes back to, the strategic question you should be asking.

I think it all comes back to incrementality, right? I think that's the thing that's failed most marketers over the last 20 years is measuring incrementality and thinking about incrementality. It's not about what you are gonna get anyway, it's about what are we getting as well as. And that, that's the thing that undoes that stupid idea about, oh, if, you reduce spend, ROI goes up.

And you go no, not necessarily." Because you could be measuring stuff that you were gonna get anyway, or you could be measuring stuff you're only getting incrementally in. If you don't do the thing that was driving incremental volume, you don't get the incremental volume. You do if you're, measuring stuff you were gonna get anyway.

James Lawrence: , Can you go into more detail on incrementality for listeners? 'Cause I don't think all marketers- Necessarily give it the attention or even understand necessarily the concept

Paul Sinkinson: Sure. If i'll paint a really extreme example, and I'm not necessarily suggesting anybody's doing this, but an extreme example would be I could bid and get loads of clicks on my branded terms on Google.

Anybody who typed in the brand name analytic partners, I could bid loads of money to get all of those people. And those people, it would then show that went on and downloaded something can... and maybe it became a sale, and that'd be fantastic. Or so the model would tell me. But the reality is that's picking up a whole lot of people who are just using the Google search to navigate, right?

The reason why the two most searched terms in the world globally are Google and Facebook are because people use it to navigate, not to search. And so you're picking up volume that you would've got anyway. The amount of volume that the model's telling you is everything that somebody, clicked on that ad, and that's because they didn't wanna scroll down and click on the free one.

They click on the sponsored one and come to you, and the model says, "This is the volume you got," but you're gonna get a whole load of that anyway. Now, if you were instead putting out a term like m- marketing analytics and driving and people didn't know who analytic partners was, then sure, that could be an incremental lead that comes in.

And if I didn't bid for that one, somebody else would get it, and that'd be volume I don't have now. So it's about looking at what are you getting on top of rather than what were you gonna get anyway. And I think a lot of the measurement that we've been doing has been claiming everything.

The reality is marketing doesn't drive everything And, it's part of the reason why we've lost some credibility in boardrooms, right?

Because we go in saying, we increase sales by 3,000%. You go there's no way. Some of these businesses with some of these results, you go, "Mate, the Australian GDP line would've changed if you did that." So yeah it's a thing about, aligning with the business and understanding, incrementality, understanding what's really driving it.

Like 5% more, that's fantastic result for a big company,

James Lawrence: yeah. I think it's also the kind of the difference between marketing measurement and marketing effectiveness.

Paul Sinkinson: Yeah.

James Lawrence: Where it took me too long in my career to If you'd asked me the difference, I would've said they're the same thing, and they're completely not.

The idea that there's a whole bunch of stuff that we as marketers quite rightly measure and report on and try to improve and all those types of things, but then how those things tie back to business metrics, hence this conversation, and we're trying to get better at it, and there's lots of things we can explain, but there's also things that are much more difficult to explain if ever possible, right?

Paul Sinkinson: Yeah, exactly that. And it's making sure you focus in where marketing is genuinely turning the dial. 'Cause I think what comes out of that then is, an incredible amount of value. It gives credibility to marketing. It shows the impact of when you reduce marketing spend and, going into tough economic times, everybody needs to be able to demonstrate that, right?

How marketing can help close the gap and the risk of reducing the marketing budget. 'Cause if you've been claiming all of these sales and your budget gets cut, it doesn't go down by as much, right? Yeah. And then everything gets thrown into doubt. Whereas if it gets cut and it goes down like what you said it would, that gives you credibility.

That makes the CFO go, "Geez, I should put that budget back up soon."

James Lawrence: Yeah. How Do you see marketing measurement I guess as an industry, is it countercyclical with kind of tougher economic times, there is more of a questioning of spend, therefore, CFOs are looking even with more scrutiny at marketing measurement and effectiveness and all those types of concepts, or not really how it works?

Paul Sinkinson: For a segment of it, definitely that reporting part of the segment where people just need to show that they're doing their due diligence. I think that part of it is definitely pressure on people need to go, "Hey, what can we do?" Then there's those CMOs who in those tough times, they wanna be brave.

They wanna really change. Yeah. And they wanna be able to grow. They wanna be able to buck the trend. And COVID Really showed that, right? There were a whole lot of people who rather than cutting spend, held their nerve and got massive gains from share of voice from cut through.

I think that there's a growing body of knowledge sitting there for brave marketers.

James Lawrence: Yeah. And I remember at the beginning of COVID, we did a lot of research and we ran a webinar ex- with all the theory at that moment in time, right? And it was kinda hard 'cause we hadn't still haven't had a technical recession in Australia through that point, right?

Yeah. For the kind of 20 years or whatever it was. So look, looking at research from the GFC, looking at, I think it was the oil crisis in the '70s coming out of the States, and I think Toyota was the case study, right? Just like hoovering up market share by investing in marketing 'cause it's more cost effective generally to do it during tough economic times when everyone else is pulling back.

Paul Sinkinson: Yeah. We saw ROIs go up by 30% during COVID.

James Lawrence: It's,

Paul Sinkinson: it's-

James Lawrence: Fascinating, isn't it? It does feel that there is, 'Cause a lot of your thought leadership has been so big on brand, right? , There's been this chronic underinvestment in brand. I think the dial, it feels like it's starting to turn there.

. I think marketers have probably known it and felt it for a long time, probably haven't necessarily felt that they've had the the conviction or the data or the white papers or whatever it might be to fight the fight. Feels that's changing. Is that what you'd be seeing?

And since we last spoke, more marketers doing a better job of having that brand versus performance conversation with non-marketers in businesses?

Paul Sinkinson: Yeah, definitely. I think there's a couple of risks that are evolving because of it. So I think , definitely far more of those chats.

What brand means and what brand activity means is part of the challenge. We know creative's really important. A crap brand ad is still crap, right? That's still part of the impact. So I think how it's executed, not just going, "Okay, cool, we're gonna do some branding," thinking that's it.

That's the start of the journey, right? That's the start of the challenge, not the end of it. But the other part is making sure people don't go too much to the other side. And it feels like we can be such extreme people in this field. We swing one way, we swing entirely, and then we swing back the other way.

And picking up the opportunity around key trading periods is still so important, right? If you're looking at Black Friday or you're looking at Christmas, you wanna be doing loads of brand before and six weeks of heavy brand activity. But then in that two to three weeks before Black Friday, particularly now, we know people are researching more before it comes, so you wanna dial up your performance then.

And 60/40, You wanna go 30/70. You wanna go really heavy on performance to pick it up because there's this huge amount of volume that comes into the market for a short amount of time. So if you spent all of your time building your demand, you assures anything. Wanna make sure you harvest all of that demand that you've built and not let your competitors do it.

We wanna make sure we don't forget when to just go hell for leather performance either.

James Lawrence: There was some data you shared last time that- , and I totally take your point, right? Like a bad, a terrible brand campaign is a terrible brand campaign. So I think we have to take for granted that we've got distinctive brand assets, it's well-conceived, we're committing to it, it's fit for category, - et cetera.

I think some data you shared was that 80% of the time a brand ad will outperform a performance ad. I don't know if that necessarily, I don't wanna put words in your mouth, but is that still holding true in the market as it is at the moment?

Paul Sinkinson: Yeah. What that comes down to I'd make sure that everybody understands it.

It's a campaign that can run for longer works better than a campaign that, can't run for as long, and a campaign that talks to more of your business performs better. So it's not that it can't be like performancy, but so many performance and price ads speak to say one product, or they speak to a price or a product feature that's only in market for, say, six months rather than for six years.

So you can still do, like brand product ads, but you wanna make sure that it's something that you can run for a long time because we know consistency drives it. We know that, once you've built the memory structure and the awareness of the campaign, the message you're trying to do, it's really efficient to remind people of it rather than have to then build a new one.

James Lawrence: Yeah.

Paul Sinkinson: You want something that you can run for a long time. That sometimes price and product means that you can't because it's no longer available.

James Lawrence: Yeah. And the idea being that if you've got 20 products and your performance ad is on one of those lines or five of those lines, even an uplift, a strong uplift there is unlikely to-

Paul Sinkinson: Yep, exactly that

James Lawrence: override the - the benefit of a smaller uplift across - all your your product lines kind of thing.

Paul Sinkinson: Yeah, exactly that.

James Lawrence: Yeah. Nice. , We spoke last time around The idea that A-an exception to that can be in the startup phase, where if you're a startup business and I think the reality is most startups, even if well-funded don't necessarily have the luxury of investing heavily into brand, and you were kinda saying that is a strong use case for your performance channels, and what they are in each category will depend.

But maybe in, in digital it might be bottom of the funnel search and, targeted paid social or something like that. When does that start to flip, and how do you see your best clients having that conversation internally where what's always worked has been performance or 90% performance and 10% brand?

How do you navigate that transition to start having the conversation, which is in order to . unlock that next phase of growth, we need to be investing in some of these kind of brand type efforts?

Paul Sinkinson: I wish I had a really good answer for that. It's a live one that we have with clients all the time and, really talented clients because the context really matters, right?

Fast-growing businesses that have gotten growth from it know that they need to change when their CPMs start to increase faster than the business coming in, right? So they go, "Okay, cool. We know we've harvested all of the demand that's out there, so we know we need to build demand."

But if you then start taking money out of something that delivers immediately for something that delivers over 12 weeks, and you've gotta deliver quarterly earnings, that's a challenge.

So how do you do it? It's probably incremental funding is the way that best practice has been done previously of not cutting performance, but adding in brand on top of it to be able to get that growth but not miss an earnings call.

With an economy going down ... incremental funding is a far harder discussion. So I think sometimes there's that, the reality of how do we get growth and how do we minimize the time transition of it. So it's not a case of going from one extreme to another. It's a case of needing to do it in steps.

Sometimes as well it's doing that by state so that you can build out the, "Hey, this is exactly how long it took with our creative," rather than, "This is how long it took with a case study."

James Lawrence: So do you mean testing it in whatever it might be, South Australia or Queensland or-

Paul Sinkinson: yeah, exactly.

I'd say more like a Queensland or a Victoria than a South Australia. That one can be tricky, scaling it. But yeah, definitely picking a state and going for it i- is far more, I think, the path to de-risk. But . it's not a case of people don't know.

I think, even the CFO knows, but it's the time transitions. So if it was a private company, a lot easier because it's quarterly earnings, right? That makes it tricky.

James Lawrence: Yeah. And I guess to that point, it comes back to the point earlier around- Stakeholder management and understanding how like it's the same as the concept around marketing in a recession, right?

You need to have buy-in on the theory. This is how it works, the different things. But to your point, I think a real live case study on your own business within a kinda controlled market versus the Toyota case study from 1973 is maybe an easier one to mount.

Paul Sinkinson: What Seems like a small semantic thing the difference of it taking 14 weeks instead of 12 weeks, that's a real issue. You know what I mean? And you need to know that. And you need to be able to tell a lot of people that and say no, don't worry, this is how long."

And it might seem like a small thing, but in terms of, reporting to boards that's a massive thing. That two weeks isn't something around the margins. That's something fundamental.

James Lawrence: Yeah. Okay. How much of your work is working with and working back to and delivering deliverables and information to marketers versus dealing with finance and or kind of leadership team, CEO, et cetera?

Paul Sinkinson: All of it goes to marketers. The ideal for us is that it goes to, Without finance, you miss the value. Like without finance being involved, you're never getting the best out of your measurement because part of that is to, one, they are key information gatekeepers you need the margin in there, but two, as budget holders and as people who tend to be in the business for longer and so can, sometimes help keep the strategy there.

If you don't have finance involved it doesn't play out. So yeah, that, that's a key part of it. I'd say 80% of the time it's through finance and senior management then as well because I think this is- One, one of the key things that's impacting the industry at the moment is this idea of real-time analytics and the need for faster and faster measurement.

I'm not convinced. I'm not saying you don't need it. What I'm saying, though very passionately, there's a whole lot of risk that people don't realize from faster measurement. I think that most marketers would agree that making short-term decisions has been at the detriment of marketing over the last five, 10 years, right?

We haven't let campaigns wear in. We've cut them too soon. We go chasing all the short-term activity. It's the very reason why the long and the short of it came out and everybody went, "Oh my God, this is amazing," because we were all doing so much of the short. Doing really fast measurement kind of gets you to more short stuff.

Like that's a fundamental issue that's playing out. And rather than that short-term measurement, because you go, "Oh, but we need to know what happens, what to do when this happens, or what to do when this happens." You can't forecast that stuff, right? Who was forecasting the war in Iraq?

Who forecasted that and had that plugged into their forecast for the year? Good luck. Who need to do that?

James Lawrence: We don't wanna get into the the speculative betting market of people betting on things that they know to happen very shortly. But I think that's right.

I take the point very validly.

Paul Sinkinson: Yeah. No, okay. Yeah. Who knew at the start of the year rather than who knew three hours beforehand? Which,

James Lawrence: which marketer or business person knew? I'll take that point.

Paul Sinkinson: Yeah. So what you need to know instead is what to do in certain scenarios, right?

And so that really is about scenario planning. And the reason why scenario planning is so much more important, I think than a prediction is that everybody goes into it going, "Okay, cool. We're expecting loads of different outcomes," rather than thinking you should come down to one. And the great thing about doing the scenarios is you...

It's interesting enough, you can get loads of stakeholders in a business together to do it, not just marketers. And not just one person running a scenario that then, a forecast that then goes up to the board to be approved, but you get loads of people going, "Okay what happens if this? What happens with this?"

And because you've got all those stakeholders, you can get everybody to endorse it, which means that you've got a plan to do under a scenario that has been endorsed, so you don't need to go and get everybody to agree to it. You don't have to educate everybody on it because everybody already was. It's an approved strategic plan that's sitting there.

So when that then appears, you've got real-time decisions. You don't need the real-time analytics, you need real-time decision-ing, and I think that's informed by getting everybody across it, right? Because what slows down decision-making? Getting people to sign it off. Getting people together.

Just 'cause you've got a dashboard that's sitting there doesn't mean that you can execute that strategy, right? It needs to go up, ... through the business. You need to get on somebody's calendar to be able to get them to approve it. But if it's already been engaged and, you've taken it through the culture of the business, everybody's across it, then you've got real-time decision-ing.

And I think that's the key thing that we should be focusing on. I think that means that, you have to have finance involved. That means you have to have senior leadership involved in a program because that's where you get a scenario that's endorsed. And marketers can then be really flexible, and they can react to things in real time because there's parameters around it.

There's agreement around it. There's, "Yep, this is what we do."

James Lawrence: It's a really interesting perspective, and I do wanna talk shortly about creative, but it's kinda how I personally feel around AI, infinite creative, the ability to spin up ridiculous volume of assets, hyper-personalization, all these things which is always a bit like the measurement piece, right?

Like 15 years ago it's all measurement's gonna be solved, and it's probably more complex now than ever before. Yeah. And I think it feels that way with creative. It's yeah, cool, you can, create a million different assets with the click of a button, but I'm not actually convinced that is gonna just solve creative and the split testing it.

But what are you testing and what are you measuring, and how do you actually connect that to marketing effectiveness down the line? I think this promise of everything just being done, I think adds more complexity and more difficulty to marketing, not less , in many ways. With that challenge around, I guess bringing stakeholders together, it's all very logical.

What are your observations around, like the audience for the podcast are all in-house marketers, right? Thematically, what are the things that you see the most effective or best in-house marketers doing within their organizations to kinda get that buy-in and get that respect across different parts of the business?

Paul Sinkinson: Thinking commercially, not thinking marketing. Being able to show that 80% of the volume's been driven by something else other than marketing. I think, building those points. I think being able to transparently show the impact of everything else. It's the challenges, right?

The sales director will always say, "No, it was our price promotion that did it." So you need to be able to show him yep, you're right. Here's your price promotion. It is massive. Yep, it's there." Ops will always say no, we've improved our processing times." So you need to be able to show, "Yep ops it's here."

You need to have the holistic measurement and show everybody the impact of their thing, because if you can show them, yep, it's in here, it's accounted for, then you can get everybody behind what your marketing's doing.

James Lawrence: I think the the point before around marketing trying to claim everything is something that I'd see in eight out of 10 businesses I'm talking to, right?

Paul Sinkinson: Yeah. And then the metric as well, right? Because if you're doing it holistically, like you can't be judging the sales promotion on engagement, you can't be judging the processing times going down by likes on a post. It means that you're pulled in then to a metric that matters for the business

James Lawrence: So yeah it's a really good perspective, right?

When we spoke last time, I think I've got a quote from you here. You were saying, "Consistency is the unheralded superhero."

Paul Sinkinson: Yes.

James Lawrence: You're talking about creative wear-out and just this obsession of marketers to put stuff into market to pull it out of market. And I struggle to reconcile that, which I totally agree with, right?

I think I, I saw you speak once and someone asked you, like what, in what timeframe do things wear out, and you just laughed and said, "I don't know because I've never seen it happen" which got a good reaction. How do you connect that which I believe to be true, and then this kind of world we're moving into where everyone's just talking about infinite creative and AI creative?

And I think that similar to your idea of real time marketing reporting not necessarily being a good thing, do we not run the risk of infinite creative, meaning that we never truly get anything in front of someone that actually holds?

Paul Sinkinson: Yeah there's a real risk that's there. I think it's different by channel.

So you know, are you gonna have infinite creative on TV? Not, probably not. For out of home, probably not. And, the risk is very much bigger there, which where consistency comes through. I think it's a question of how, right? Because it's this challenge of wear-out on social, right?

But it's not wear-out from people, it's wear-out of the algorithm. So you need new creative coming in to, to make sure that it works. But does it need to be a new creative platform? No. From what I'm seeing, it's these small little tweaks, right? It's the framing of the shot, it's the colors.

So you're keeping your distinctive assets, you're keeping that same campaign platform. You're just doing little edits on it. And I think that might be one where these infinite edits come through, because if you've still got your long form brand idea and, you've worked that through. And I think as we move into AI guided search as well, it's gonna be even more important because if you've got your core creative idea, that can go long.

And by long I mean into your website, into all of the content that you do, because then that's getting you into more natural conversations, which is gonna help you in AI search, right? So I think that helps you go long. But then if you've got that understood really well, it gives you the framing for loads of social creative to come out that you can be testing that can also be playing on topical things so you're not burning out a creative team, trying to chase the latest trend that's coming out.

But you need it really locked down. You need really tight parameters on it. And what I'm seeing, the best stuff is just subtle tweaks on it, not letting it go too far.

James Lawrence: Yeah. Interesting. So we run many campaigns, paid social, right? And , to That point, you get fatigue in terms of algorithmic fatigue.

Paul Sinkinson: Yeah.

James Lawrence: Which I think is, it's interesting to look at that distinct as opposed to creative wearout, right? So if you are take the spec, not that we work with them, but the Specsavers should have gone to Specsavers, the idea being that the core campaign is still there, right? The theme is there.

Yeah. And you see the application of that in many ways, right? Welcome to Melbourne, welcome to Sydney and the airport, the the billboard over the Anzac Bridge with, it's, we're still connecting with it. We've still got a lot of those distinctive brand assets tapping in. So the idea of using, When you have algorithmic fatigue, but still working within the same kind of paradigm, right?

Paul Sinkinson: Yeah, exactly. And it- it's about understanding where to put the boundaries.

James Lawrence: When you started to talk there you talked about some of that out-of-home stuff, TV street media, et cetera. Is that where you were saying fatigue is more of an issue in terms of people- No ... taking it out of market too soon?

Paul Sinkinson: Yeah. People definitely take it out too soon. But there, like that's where, we kinda see four years and it's actually still getting better rather than it's declining. It's this concept of wear in.

James Lawrence: Yeah, okay. And so the idea is it's okay for Specsavers to start playing around with welcome to Sydney, welcome to Melbourne, as long as the us as potential customers are still connecting it through to that one core kind of campaign?

Paul Sinkinson: Yeah, it's the same creative platform, right? Should've gone to Specsavers.

James Lawrence: Yeah, interesting. And then how are you guys, how do you measure that? How do you measure when, wear in and then in the, very unusual case where you actually do start to see wear out? Like, where, how does that kinda work?

Paul Sinkinson: So creative goes in as a factor into the models, so you can, kinda strip everything else out. You can put in creative. You can put in creative pre-testing as well, to be able to see, okay, how much is, the creative itself kind of driving in the play there. But you're looking to be able to see at what point does it either tarp for tarp, impression for impression, view for view, does it stop giving you the same lift as it did before?

And it's not just, we don't say at the point that it starts to give you less of a lift do you cut it out? Because at what point does it give you less of a lift that it could warrant, it could recover the cost of making a new campaign? 'Cause we think it should financially wear out, not just wear out.

James Lawrence: And the cost of a campaign being the cost of the creative or not the opportunity cost of then having to start to build it in again.

Paul Sinkinson: Yeah, exactly.

James Lawrence: Yeah, interesting. What creative are you seeing out there that's actually resonating at the moment?

Paul Sinkinson: It varies significantly, right? Things that talk to value are doing well. That's not things that are talking to price. Things that are talking to value content-wise works really well. And, as Is always anything with humor- ... is still cutting through really well. And then it comes down to, how you express it and where you put it.

There's ads that are humorous and laddish that work well on targeted media where it's hitting that audience, whereas if you try to put it on TV, it'd just piss people off. It still comes down to what content for what channel. You can't run, all content on all channels.

James Lawrence: Yeah.

It makes sense.

Paul Sinkinson: Yeah. But yeah, back to Specsavers, right? Something that's got a little bit of wit, a little bit of humor to it.

James Lawrence: Yeah. We had Josh from Oh! Media who's been on the pod before, but like present to our just our internal staff meeting and just- Talking around how, I think outdoor media is very heavily skewed towards brand building, right?

Versus kind of being a performance media channel. Yep. And then just the way to connect emotionally. And it's also, it's an interesting media in the sense that unlike watching TV or being on Facebook where you're distracted and have to deal with it in some way, we cut out all outdoor media unless there's something about it that makes us actually engage with it.

Yeah. So it really does have to have that element of grab, and more often than not it is humor or really awesome contextual kind of, you're next to the garage and you've got something related to that, or next to the pub that kind of breaks the ice a little bit.

Paul Sinkinson: Yeah, I think if people started thinking about their out-of-home creative it'd be awesome because so much of it's an in frame of the TVC or a pack shot, right?

James Lawrence: Yeah.

Paul Sinkinson: And it needs to be there, but we've done some work looking at neuro-tested creative on out-of-home, and it was, like, three times as high ROI. And when you look at what part was put onto it, you didn't need to do neurotesting on it to be able to pick it. I think everybody would've been the creative director, but the neurotesting gave them the confidence and the intestinal fortitude to do it.

Yeah ... so it's a case of going what part unlocks the emotion and is gonna, kinda come to memory and get that reaction. I think if people were doing that rather than just the end frame of the TVC or a pack shot, we'd be winning.

James Lawrence: Yeah. It takes you gotta be kinda brave 'cause I've talked to him personally and I was like, "You, I'm sure you just know when it's a winner, right?"

He goes, "It's every time." And it's the one where it's they're trying to cram 20 words onto a billboard and trying to convince everyone of the brand and the performance, and it's just like no one cares, right? And then he'll show the Barbie billboard, which is literally just a pink billboard with the Barbie pink and then the date in the Barbie font, and you kinda know there must be a Barbie movie coming.

And even though it's the billboard couldn't have said less, right? But to be brave enough to do that, it's you kinda gotta zag when everyone's zigging.

Paul Sinkinson: Yeah. And also just to think about the platform. People are still putting TV ads on YouTube, right? And it means that somebody's still seeing an ad that's designed for a big screen on their mobile.

It's those vistas, right? Think about a watch ad or a car ad where it's, across this landscape, it's absolutely fantastic, blown up even better on a cinema screen, and it's brilliant. You watching it on a mobile phone and it's four pixels is what you can make out.

It's what the hell is this?

So there's still that sort of a thing going on rather than going it needs to be tighter frame shots, it needs action faster, it needs to be visually violent that an old man like me goes, "Oh God, what's going on here?" And my kids go, "Oh, this is awesome."

James Lawrence: Yeah, just it makes sense, right? It's very true. I wanted to wrap up with some rapid fire predictions. Sure ... I thought that would, might be a fun way to, to finish it up. Let's go through. Let's see how we go. So two years ago you were optimistic but cautious on AI.

Where are you now?

Paul Sinkinson: I think it's delivering better two areas. It's helping pull together insight. It's doing great things around automating bollock tasks. But what it's not doing is the thing that I want it to fix, which is data.

James Lawrence: Okay. Let's, I'm gonna check back in two years with you.

What's the most overrated trend or fad in marketing measurement right now?

Paul Sinkinson: Faster.

James Lawrence: That's good. Most underrated?

Paul Sinkinson: Getting everybody together, so being strategic

James Lawrence: One, one thing marketers should stop doing?

Paul Sinkinson: Not measuring incrementality, so last click

James Lawrence: One thing marketers should do more of?

Paul Sinkinson: Getting finance involved

James Lawrence: What will matter more in five years than people realize today?

Paul Sinkinson: Investing in their brand and building some auditory distinctive assets

James Lawrence: Auditory?

Paul Sinkinson: Yep.

James Lawrence: Can we go into that? We didn't, we haven't talked about that ...

Paul Sinkinson: yeah. I was having that very chat this week with people going yeah, we've got a sonic asset." And you go no, a sonic asset gives you half of what a jingle does from on the basis of a distinctive asset."

It's the dirty jingle, right? But if I told you that Aussie kids are... and it hasn't been on air it- it's come back, but it's been ages, right? You can hum the McDonald's jingle to people and there's 30 years of brand value.

James Lawrence: Yeah

Paul Sinkinson: In your head. The jingle's the number one thing.

I think with TV viewership going down it's more and more important to be able to link all of your creative, and as we said, all that creative's gotta be, for a channel. The stuff going out on social you need something that's gonna pull it together. What's the one thing that can?

What's the thing that can go on a brand ad, and a price ad, and a product ad, and across any channel? The jingle

James Lawrence: Yeah. That's great. But I didn't stay true to the rapid fire, but I wanted to dig into that. What belief about marketing will still be true in 20 years' time?

Paul Sinkinson: That it's the growth driver.

James Lawrence: I love it. Paul, thanks for coming back onto the podcast.

Paul Sinkinson: No worries. Thanks for having me.