Building the Agency Operating System | Adam Woods & Amy Crowther | Signal & Noise LIVE at Advertising Week 2026
If AI can execute in seconds what agencies once billed for in hours, what should clients pay for next?
Filmed live at Advertising Week 2026 at MCM Studios in New York City, this episode of Signal & Noise features Adam Woods, Global Chief Product Officer, and Amy Crowther, CEO, Americas, at Incubeta.
They join hosts Rio Longacre and Brett House for a candid conversation about the future of agencies, the economics of AI, and what it takes to connect technology, talent, and data around better business outcomes.
Building an agency operating system means more than assembling a collection of tools. Adam and Amy discuss how agencies can bring creative, media, measurement, and client-owned intelligence into a more connected way of working—and why changing how people work is just as important as changing the technology. From creative production and data infrastructure to client relationships and pricing, this conversation gets into the practical decisions behind agency transformation.
In this episode:
Beyond billable hours: How AI challenges staffing-based fees and traditional service models, and why the real opportunity is in how agencies reinvest efficiency gains.
Connecting creative and media: Why clients still struggle with the gap between creative ideas and media execution, and how shared performance data can give teams a stronger starting point.
Creative intelligence in practice: How understanding which assets work—and why—can inform production, testing, and optimization, rather than leaving creative teams waiting for feedback.
Client-owned data and intelligence: Why brands should build on their own knowledge, how fragmented tools create disconnected data, and what agencies can do to help clients retain control.
Measuring business impact: Moving beyond click-through rates, cheap impressions, and platform-reported success to questions about profitable growth, customer acquisition costs, and incremental contribution.
Bringing brand and performance together: Why short-term results need to be understood alongside long-term brand investment, and how agencies can help CMOs make a stronger case to CFOs.
Making transformation practical: Why the starting point may be as simple as getting a brand’s creative assets consistent, proving value, and building momentum from there.
Rethinking the value exchange: When to challenge an RFP, why trust matters, and how outcome-based pricing can work when costs, incentives, and shared goals are realistically aligned.
Adam brings a product and technology perspective to building connected capabilities. Amy grounds the discussion in the realities of leading teams, educating clients, and turning those capabilities into work that makes a difference. Together, they make the case for an agency relationship built around informed decisions, shared accountability, and measurable value. AI expands what is possible; the harder work is deciding what to build, how to deploy it, and which problems are worth solving.
Watch or listen, then tell us: What should the agency of the future be accountable for—and how should clients pay for it?
Subscribe to Signal & Noise on YouTube or follow on Spotify for more conversations about data, technology, AI, and the people shaping advertising.
#SignalAndNoise #AdvertisingWeek2026 #AdWeek2026 #Incubeta #FutureOfAgencies #AgencyTransformation #CreativeIntelligence #AgenticAI #MarketingEffectiveness
Transcript:
speaker-0 (00:13.814)
Lights, camera action, this is Brett House live at Signal and Noise Advertising Week 2026, joined by my co-host Rio Longanchor. And we've got special guests in Adam Woods.
The Chief Product Officer of Incubita, and Amy Crowther, the CEO Americas of Incubita, who lives up in Woodstock, right, which is a great lifestyle choice. We've been talking a lot about lifestyle. Community, yeah, that's an awesome spot. welcome to the show. today we're going to talk about sort of the transforming agency, the agency of the future. Not to be hackneed or trite about it. but for those that don't know Incubita, very large outcomes based marketing agency that was established.
quite a few years ago but has been has grown rapidly across what twelve countries? Twelve or fifteen countries globally. in r and Adam was recently brought on as the chief product officer. I don't know if it's so recent. Yeah. If you start to add up the total time work to it probably feels like a lifetime.
speaker-1 (01:01.368)
Yeah.
speaker-1 (01:06.83)
This year. Yeah.
speaker-2 (01:12.414)
He's based in Denver, you should call it
speaker-1 (01:14.094)
Right.
speaker-0 (01:14.574)
And and what and and I work for Incubita, but this conversation will be fully open and transparent. There's no advertorial sales salesmanship around the Incubita brand.
speaker-2 (01:27.086)
We got some great sponsors for this event. Outstanding sponsors. Main sponsor being Media Radar. Thank you, Media Radar. Also Greedyal, PubEx, Nexa, and Telium. So some outstanding sponsors. We couldn't have done this event without them. So thank you.
speaker-0 (01:40.62)
Yeah, I mean we were saying we went from a very small press room corner, filming on an iPhone to like a Hollywood production facility where John Hamm is gonna be a a floor above us for anybody that hasn't watched Mad Men, if you're in the advertising industry. We have met some people that haven't watched the show.
speaker-2 (01:46.104)
Yeah.
speaker-2 (01:58.552)
But it is the definitive show on Madison Avenue.
speaker-0 (02:02.746)
It is just addictive. I watched it in like two months of just like every night, like at nine PM. I'd watch one one episode before I went to bed. But but back to Incubita, you guys are leading this this transformation of of both people and and tech within the organization. You're an independent agency, right? It's obviously a lot of news in the in the agency world and space has been around the operating systems coming out of the hold codes and a lot of fast moving you know, indies trying to really build
tech and kind of processes and and people alignment around kind of what is next really, right? From from a productized solution.
speaker-2 (02:39.978)
Well yeah, but I think it seems like we're at a big inflection point. This isn't the first time the industry's gone through this, right? You know, every seems like every couple of decades things are changing, technology's emerging, and everyone will say, let's the agencies they're done, they're in trouble.
speaker-0 (03:06.252)
Yeah, totally. And and and we had Rashad Tabacawala, great episode if people haven't listened to it. He's a he's a he's a a Titan in the industry, chief strategy officer at Publicis for 37 years. He probably would hate me for saying that he was there for almost four decades. Right. And he said organ people change organizations.
speaker-2 (03:15.896)
Phoenix.
speaker-2 (03:26.786)
Don't transform, people do.
speaker-0 (03:28.214)
Yeah, don't transform people do. Thank for the I was gonna about to mess that one up, right? And and Adam, you and I were kind of talking talking off off air about that topic, right? And and you guys are certainly leading a people, talent transformation around what you're building from a tech perspective. So tell the audience a little bit about about
speaker-1 (03:46.144)
Yeah, yeah, I mean so so we've been really kind of focused on like how do we as an agency adapt into this new world and and you know it really is like a lot of the traditional agency levers of how the businesses ran, how we made money, things like that, have come under pressure and and have, you know, wildly different models available to us because of the advance advancements of technology and AI specifically.
speaker-0 (04:11.475)
And models are we talking business models or agentic models?
speaker-1 (04:14.606)
Talking about like business models but powered by agentic models. Right. And so so you know, we've really been looking at it as, you know, there's this big inflection point in the industry. And you know, for us as as an independent agency, that actually gives us a lot of opportunity because we can move extremely quickly. But to your point, it is about, you know, how do we change the way people think about their work, their jobs, how they engage, what they do. but
also the business models that we work inside of. So, you know, there's there's a lot to the transformation and it is it's not just a company thing it at at an individual level, it is a very, you know, truly personal individual thing of like how do you how do you bring this new technology and use it in your day to day across like a wide range of things.
speaker-2 (05:04.012)
How are those two related? I'd love you c if you could maybe unpack that a little bit. How is the actual the work and what people do related to the business model? I think s I yeah, I mean I think that's the question of the day. Maybe if you could start there.
speaker-3 (05:15.34)
Yeah, mean, it really is about change management internally as much as it is externally with our clients as well. think I agree with that sentiment that it is about people that transform and not organizations. think helping them understand where the incremental value can come from in their day to day by using technology, not just doing it for the sake of it and just because it's there, but where does it really free up time and capacity and where can we do things better to a better outcome and quicker at the end of the day? And I think
We're doing a lot around our teams and people. There's the training aspect of it and make sure they understand what's available to them and how to deploy it and not duplicate effort. But then really how do you put it into practice in a way that is understood the value outcome for a client? And I think we're trying to install team members to understand it's not just having your hands on these things. If you're not using it appropriately, it's...
you haven't understood it in the right way. And that's really, you have to be accountable for like, every one of us has to be accountable to keep pace with the way that things are moving and keep taking advantage of the opportunity in front of us. think, you know, you mentioned the amount of times agencies have died and come back like a cockroach of the industry, but that's why we work in agency, right? Is because we constantly evolving and changing. That's what gets us out of bed. That's why we exist. And so it's a challenge, but it's also the most fun part of what we do.
speaker-2 (06:41.826)
Looking at the FTE model, there's been a lot of discussion over the last couple of years. I mean, it's it's impacting every industry in terms of what we look how we look at work and how organizations justify the things they do and they build clients for, especially in the services industry, right? So buildable apps. Yeah, like you look at consulting, for example. There's been a there's a ton of talk consulting about how do justify time and materials contracts when a lot of the work that you're billing people for can be done in
seconds that took days or weeks even before. How do you justify that? And how do you, you know, how does that potentially even change the all structure of the organization itself? So that being that being one example, it's looking at agencies specifically, there's been a lot of talk. We had some good a couple of good episodes when we talked with people about is the FTA FTE model being challenged? And if so, like what does that mean for an agency and how are they responding? Love if you can maybe talk about that a little bit, Adam.
speaker-1 (07:34.092)
Yeah, yeah, I think so a couple of things. Yeah, you know, I think that it it really puts the onus on us as, you know, that that service provider, you know, the that's creating that value to to do a little bit more to think about like what are we doing that actually creates value. and so, you know, when we think about that from a product perspective, there's a lot of places where we've found that those efficiency gains, right? And and we all we all know the
Examples of it, right? Like there's you know things where we can do you know elements of creative production where you know maybe we're like selecting images and tagging them at a rate that by hand was a very slow activity, and and now we can you know sort through hundreds of thousands or millions of images and figure out which ones are going to be the best pairings. And you know, that we can do those things and like yes, there's time that came out of the overall program, and that we can do things that much faster, but then I think that.
Where the innovation lies and where you know where the value is is okay, how are you reinvesting that savings? And you know, in one of the cases with a retailer that we work with in the UK, the reinvestment has been around the the volume of creative production. So where before it might have been you know like 20 or 30 or 50 creatives a week, you know, now it can be hundreds or you know, like even really thousands. and then you start to get to being able to actually go.
much deeper with the platforms and putting thing in flight things in flight and really understanding that like, you know, where that acceleration happened actually gives them more benefit and more performance. And then how do we structure the relationship around that?
speaker-0 (09:20.054)
Yeah. Now now so that's an iterative process, right? We certainly think that AI can take over sort of the iteration, the versioning of ad creative as a perfect example. It could be content creative. But but does that free up people's time? Like you you know, you run client relationships, right? Does that free up people's time to add more value to do higher value strategic work work? It could be the big idea from a creative perspective.
speaker-2 (09:42.86)
That one big idea, yeah.
speaker-0 (09:44.088)
Right. you know, where human sort of judgment and creative sort of power comes into play, something that the AI really can't do 'cause it sort of builds off the median more or less, right? So how how do you guys think about that in terms of freeing up time for higher value work?
speaker-3 (09:58.606)
Yeah, I think it kind of, changes the starting point for the teams. Honestly, it's, you know,
I was on a panel a couple of weeks ago about the topic of data and creativity coming together and my God, what does that mean? And data's always been in creativity. It's just not been used at every element of the development process, right? It's always been strategies to use data to guide insights and human understanding and then they build that into the creative idea and then you execute and then way down the road, maybe you measure if it worked or not, right? It just now, the fueling of data in, it's in every part of the workflow. And so, you we don't...
create new concepts of creative until we've looked at what do you already have? Is it working? Which bits of it are working and why not? And really understand what it is about the brand that's distinctive, that's making it perform in media, and then we build. And so the teams, think, it's less about having free time to be more strategic and creative. think their starting point is a lot more solid and built on foundation. A lot more informed.
speaker-0 (11:01.482)
It's a lot more in really working in we had a really good conversation with Alex Colmer of yeah
speaker-2 (11:02.334)
Four.
speaker-3 (11:04.512)
And more interesting conversations with clients.
speaker-2 (11:10.006)
We did. A creative intelligence.
speaker-0 (11:12.692)
intelligence and he had a video mob and he's sort of the board mob sorry the board and founder you know I don't think he's the acting CEO and he talked about how it's almost criminal how creative intelligence like the insights that go into creative tracking in in whether it's real time or not has been almost criminally undervalued and and ignored right all the work was good has been going into audience targeting and audience fidelity everything that's driven a lot of the media ecos the the programmatic media ecosystem
speaker-2 (11:15.906)
Fit mob, yeah.
speaker-0 (11:42.576)
ecosystem, the digital ecosystem, but not enough is going into like, well this is what actually has the largest impact on consumer attention, engagement, conversion. Why are we not why are creatives still complaining about not getting any feedback?
speaker-2 (11:55.486)
Yeah, it's interesting. H and his argument too, which I do I do agree with, was everyone has access to the same large language models and the same tools now. So the differentiator really is not gonna be that. It's going to be creative and in order to have better creative and insight you need better insights and creative, hence the importance of creative intelligence. I wonder do you agree with that state?
speaker-3 (12:16.558)
100 % and I think it's one step further than that, it's how you deploy it and how you bring the teams together talking on the topic of agency. know, media agency historically, especially holding company, they don't want to play in that space, It's about the media, it's about the environment that the creative shows up. As long as they've got the right asset with the right aspect ratio and format, they're good to go. The plan goes ahead.
That's their KPI. That is their KPI and the outcome of that. Creative, equally so. The best idea, the award-winning creative, know, make sure they don't really want to spend time in the production and making sure it's fit for media. The messy space in between that clients have been in and having to stitch the two together and is this hand talking to this hand. That's honestly where 80 % of our work is today. And it's building on the creative intelligence, but helping...
know, stitch that all together because you make everybody's work work better. Two weeks ago, you know, in New York around, you know, one of the big C creative agencies in their HQ with a client, with the holding company media agency, with the creative team, all looking at the same data. How do we make every ad, every bit of media, every performance element work harder? That's the new conversation.
speaker-2 (13:30.83)
Interesting you say that too, 'cause I mean for those who are not familiar with the agency model, there was this decision, I think it started in the eighties, right, to break off the media arm from the creative and another arm, which at the time, I mean, maybe it made sense. I mean, I was before my time, right, in in agency world. But, you know, the argument was d it can be more efficient, it can be better economies of scale, they can you know have better specialization. But I think the problem is it broke that link you just described of
You deploy the creative, you wanna see how it performs actually out in the wild. You wanna have it actually attached to your media and your measurement and you wanna be able to have that closed loop, but how do we improve it? Right. So I think that contributed to this like breakage between the actual creative itself, the insights in a creative and in the media spin and like where it's where it's deployed a lot of the time, right? Yeah. And I think I I'm seeing a trend towards like restoring that now. Certainly in the big hold codes are trying. I mean it's I think it's harder for them though.
speaker-1 (14:23.63)
Yeah. Yeah, I mean I I think that we like we we clearly see that convergence as well. And it you know a lot of kind of the product and technology that we've been working on, you know, we we have called are calling it seamless by incubata. And it really is that notion of like how do you bring these things back together? And you know, a really good example of it is kind of that marriage of creative and media. and so we we have capabilities, you know, solution on top of product where we look at
Creative, we break it down, you know, really kind of like at a very granular level, scene by scene, element by element, you know, looking at the transcript, looking at all of the things that go into that creative, and then you know, finding those things, and then looking at performance of media in market, and then trying to correlate elements of the creative. So not even just like the creative itself, but decomposing it all the way down to you know the elements inside.
of it and then looking at like okay you know first iteration how can we recompose the creative or how can we build a better creative that comes from data and measurement and and is really fueling that creative process but then as we put those into market and then we look at that as an over and overtime improvement it's like yeah how do you create a flywheel that is creative and media together informing each other and using that you know to to get to this notion of like
continuous improvement over time. So you know, never just a one off, never just like hey I you know, I helped a brand make a really incredible piece of creative. It's like no I want to actually help them make years worth of creative that that continuously improves because of those learning
speaker-0 (16:09.046)
Yeah, and is that how controlled is that process, right? So you've got a you've got a creative brief, you've got let's say a creative master template with a bunch of different components that can be switched and swapped, whether they're product mix or messaging mix or offer mix, whatever color, location, you know, all these things that that might speak to the cohort that you're targeting. How do you actually do you automate the process of optimizing all of the components? Right? And how do you control for things like hallucination, things like, well, hey, it's it's it's now in market and it's not appearing the way we expect it to be.
speaker-1 (16:30.359)
Yeah.
speaker-0 (16:39.01)
period like breakage in the in that sort of automated process.
speaker-1 (16:41.388)
Yeah. Yeah. So so e a hundred percent that's what we're doing. And I think that, you know, this this is where kind of like some of the subtlety and complexity comes into play, which is, you know, it's not about, you know, giving an LLM a brief and saying, like, here, cut this whole commercial like front to back, right? Like that that doesn't work. you know, and a lot of brands yeah and and a lot of brands are you know like that that's not what they want, right? But it's how do you create a program that
that kind of like encompasses the entire problem with AI, where you can look at like step change and elements and you know still retain a lot of control over it. So you've got your brand guidelines, you've got your templates, you've got elements, you know, and and even you know using a lot of their, you know, like their photography footage, their video footage, previous creative now. Yeah.
speaker-0 (17:34.25)
for your kind of built-in governance of show and how. As long as you build that context well enough, that becomes part of the kind of the brand IP, right? Like this is how we defined our brand and how we how we how we market. And then you've got a level of control over what's actually being automated from an optimization perspective.
speaker-2 (17:51.998)
Looking at the automation, I think this is almost the irony of AI for agencies, right? Is in order to win business, agencies need to say, we're investing in AI, we're AI first, we're AI forward, we're training our people. And I mean, everyone's saying it, so in order to in order to even be in the game and to win, you need to say that. Yet the yet at some point, and I'm wondering, I'm I'm curious what you're hearing from your clients, is I guess my question, at some point the clients are going to say, okay, you you're blowing all the smoke about AI.
What does it mean for us? I mean you're doing your work faster. Are you doing it cheaper? Are you doing it faster? Are you doing it better? Is it gonna cost are your fees gonna be lower? Like what are you hearing from clients and how are you responding?
speaker-3 (18:32.14)
Yeah, I think it's broader than AI. think it's the question of all agencies, which is not what services are you selling me, but what outcomes are you solving for me? And that's got really messy because it's no longer just you're buying my media, you're my creative agency, you're my measurement partner. They're in this world as well, trying to, they're being asked by their CEO and CFO, how are you streamlining your marketing team using AI and becoming more efficient and more effective? And so it's a real challenge moment that
we're at and I think that's why we're so heavily leaning into the outcomes in both our pricing and our solutioning because it's not clean cut anymore and I think that from an agency definitely our size and being independent allows us to do this a lot quicker is just operate in that messiness. mean the reason we call it seamless by incubator is because it's the only word that makes sense of kind of what everyone's trying to achieve right now. There's a lot of friction.
speaker-0 (19:29.214)
this process right and that's where that's where startups come in and play right whether it's uber in the taxi industry there's a lot of friction to ordering a taxi if you live in the burbs right you know you get a broken down car that shows up an hour late and they remove friction from that process and suddenly you've got a business model that's right isn't that isn't that a similar concept
speaker-1 (19:45.932)
Yeah. Yeah, but it you know, I mean there's there's also kind of like a trap in, you know, the the ease of how fast people can build right now. Yeah. Right. And you know, one of the things that we've seen when we look back at like you know at like the Loom Escape days, right? It was like a hundred and fifty companies in Martech and you know, you look at that the chart and you're like, my god, our industry's so complex, right?
speaker-0 (20:09.708)
Well, fifteen thousand Martech companies, according to Scott Brink today. And that started if you go back fifteen years ago, fifteen, twenty years ago, it was about two hundred and fifty three hundred
speaker-1 (20:12.856)
Yeah.
speaker-1 (20:18.178)
Yeah.
speaker-2 (20:18.506)
Yeah, then you add an adject.
speaker-1 (20:23.52)
Yeah, and and you know, when you think about that, and it's like, you know, if you're a brand, the more and more of those point solutions you bring in, the more you're fragmenting your estate and your data, and the less things are working together in that like seamless way. And so, you know, a lot of what we think about is like how do we compose the technology that's our own tech that you know that we're getting from partners, that we're using with platforms, and really composing that in a way that it's like, yeah, all of this data actually.
actually does work together and that you've got a single data spine and it lives in your environment.
speaker-0 (20:59.742)
You have sovereignty and control over that data. Now the AI topic, but but I think it it it points a a good picture of of really AI is not a strategy, right? And I think Rashad said AI AI is about as much a strategy as electricity is a strategy. Meaning, well, if you're competing against kerosene lamps, right, you know, yeah, maybe you've got it one up. Because if everybody's got access to the same stuff, to the same predictive capa capacity and compute capacity, we're we know that data is l largely, especially third party data is largely the same across the
Right? So, how do you actually get differentiated, right? And part of it's about building that that IP on top of your data infrastructure and kind of owning a lot of what's what you know, which is the context layer, right? and that ties into this topic of well, one, would you agree that AI is not a strategy? And two, how does that tie into things like AI sovereignty, like having control over what your AI is doing that's that's basically being trained on your brand, on your performance results, on your data.
It's very specific to you and it's not being shared across a a number of other
speaker-1 (22:03.534)
I mean i a I agree with you, like it like AI is not a strategy, it's a buzzword, right? It's like we all look at it and we say like, this is gonna transform everything and you know, like it's it's a it's like a true statement that's also entirely empty. Right. Like what do you what do you do with that, right?
speaker-3 (22:17.805)
How you respond to that?
speaker-2 (22:18.998)
Brand saying because like you think about it, they AI sovereignty. So brand owns like like their their data, it's theirs, you know, there any any any outcomes from that, any performance in that in theory, those are those are theirs. And the agencies obviously acting on behalf of the brand. So but on one hand you could argue the agencies are are very well positioned to succeed in this because they can aggregate data and insights from all their clients, and they can use that, you know I mean they they can not only buy
be able to buy more media and the argument is they can get better rates, but also they can like if if having better AI means having better data to feed the AI, agencies should be in a great position. But how do you square that with the concept of data sovereignty and brand zoning their own data?
speaker-3 (23:01.55)
Yeah, I mean, it's always been a challenge, the walls that you have to put between different client data paths. think, you know, we very much advocate client-owned and don't rent intelligence off anybody else. that's the most exciting thing for brands today. And I think you think about the years and some established brands, the years of data that they've got around media and, you know, as a starting point to start to build on applying a layer of a knowledge graph or an AI infrastructure on top of that and making it make sense
for the business, that's not exclusive in any way to any agency or any brand. And I think we just do not own our client data as the way to, I think that the argument for aggregating media buying and the cost of that, again, those things are not as important as they once were. I I was talking at a session last week about, I think there's agreement across the industry, media buying and execution is going to be one of the first things that drops off the radar.
because clients can own that themselves now. It's the intelligence that they have around about where they buy and what they do.
speaker-2 (24:06.356)
That is the new role. last week, which was the Amazon ads event, there was was one of the keynotes, this is wild. So they had a guy from one of the hold codes up there. I felt bad for him because the Amazon pre approves everything you're gonna say. Like you don't really have a lot of choice. And he had to say and Amazon's you know, they got rid of the DSP. They're calling it like ads agent now. Yeah, right. Yeah. And what's crazy is
They had this guy get up there, and like the whole Amazon pitch was kind of like the meta pitch from the Zuck said a couple years ago. You tell us what you want to sell, you tell us your budget and when you want it done by, and everything will be automated. It's kind of Amazon's pitch is you tell us what product you want to sell, tell much how much money you want to spend, you put it in the machine, and we'll start cranking this guy and this agency guy got up there, and he was just saying how basically we're gonna be out of jobs because this it's gonna be doing everything, right? So I'd love if you could respond to that. I mean, do you agree with that? Is that is that aligning with what you just said about like media buying the actual
speaker-3 (25:02.562)
Yeah, mean, it already is. Using Google's platform, using Meta, these things are pretty, they're already automated and our teams have already shrunk over time that do that work, but it's moved into strategic planning and being more consultative with clients. So it's changing and that's just going to continue to pivot. I think from a pricing perspective, the value that we unlock as agency, the role of agency has always been to, you
help clients navigate things and give an external perspective. That's not going away, it's just going to be how we do it. And I think we need more clients to be asking that of their agency partners and not stuck in the old ways of FTEs and percentage of media engagements. That doesn't allow for that to happen.
speaker-0 (25:49.154)
Yeah, and it's getting out of the sort of work that again can be automated. It's the creative production work. It's the yeah it's the it's the data wrangling and trying to pull all of your data together and get insights from it that are
speaker-2 (25:59.362)
Just in platform like stuff. It's just a bunch of like you the twenty two, twenty two, twenty two year old thing about how is media actually like executed as twenty two year olds working twenty two hours a day for
22,000 dollars a year, that was the old joke, right? So yeah, maybe that goes away, right? Maybe or maybe you don't need as many twenty maybe it's thirty-year-olds, but you're paying them a little more and there's fewer of them. I don't I don't know. But but I it's interesting too. I mean, I saw this, there's been a trend, I guess, on the media side where some of the big brands there was that famous RFP that one of the OEM auto OEMs issued a few years ago, where I think they basically said like no mi no commission, because the normal commission you know back in the day was like fifteen percent. No commission with you kind of basically saying you'll find your money, like whether I mean I'm gonna I'm assuming that was the rebate.
or through you know principle based buying, but like it just I just thought like that's not an informed way to buy, right? I I couldn't b like and I've seen more and more of that as a trend. Is that what you're seeing? And they were thoughts on that?
speaker-1 (26:52.044)
But but even I mean even at that like you know a lot of times like some of the more sophisticated brands are starting to push for audit rights and looking for those you know those remit remate agreements in the back end and like you know pushing for a lot more visibility which is you know one of the things that that we we actually talk about and think about a lot is like you know there should be transparency in media, right? Like you should understand. Yeah. And and I think that like
speaker-0 (27:18.84)
What what is the yeah, what is the quality the real quality? Like if you get a quality score for the media that I'm purchasing, where is it appearing? How is it appearing? Is it appearing at b you know, there's all these the markup
speaker-2 (27:28.364)
What are the markup?
speaker-1 (27:29.274)
Yeah, what are But but then but then you also start to reframe that conversation and think about like how do I actually help a brand get to the outcome, right? And so, you know, it may not be about like what's the actual like quality and the volume and the CPMs, right? Because those are you know, those are always kind of like secondary things. You know, at the end of the day, like every brand that I ever talk to, right, like their goals are I want to grow and I want to grow in a capital efficient way, right? Like I
need to acquire customers for less than than I make you know from a contribution margin off of them. And and so when you start to think about that, then it's like, okay, how am I treating my media investment as you know something that I can really truly justify to my CFO, to my board, and that it's you know it's this holistic outcome. And it's not, you know, it's not like over focusing on lower funnel and like doing like CPC kind of stuff all the time. Yeah.
speaker-0 (28:29.07)
Yeah, it's a balance between brand and performance. But I but I would I would argue that like counterpoint that that media quality is critically important because that's how much working media is actually going. You know, right now you know it might be it might be 50 cents to a dollar that's actually working media, especially when you go in broad principle-based buying where you might be advertising on, you know, a thousand and one AI generated recipe sites, right, that have a t a huge ad density, right? Your ads appearing below the fold, so it's not really effective.
speaker-1 (28:29.148)
Yeah.
speaker-1 (28:36.355)
yeah, a hundred percent.
speaker-3 (29:07.682)
That's the one way to measure. Yeah, I measuring by outcome. That's it. That's the only thing, the standard that we can hold ourselves to. think there are businesses like IAB and others that have done a really good job of embedding some best practices in there. But ultimately, if your agency is not, it still gets me. Sometimes we look at click-through rate and I saw something the other day and it was like, my God, we've like 20 times better click-through rate. And it was like 0.0025 to now zeroes.
How is that? What marketer is walking into any room and claiming that as a success metric, you know?
speaker-2 (29:37.336)
Yeah.
speaker-2 (29:46.134)
Or even agencies like that are told you will win this bid based on your CPM rates with major pubs. I just think that is like that, okay, then you're guaranteeing you're gonna be buying the lowest cost remnant garbage invent like inventory on like recipe sites and other MFAs. I just think and but a lot of RFPs have that, which I think which I think is a problem. So how much of this is a need for better informed
Well guess my question is how much of this is better informed buyers versus let's say agencies actually doing the right things? Or you think you think we need both? How
speaker-3 (30:20.238)
What is it? I think it's both. I think it kind of brings us back to the FTE or pricing conversation of how you engage with an agency. think when you do have an RFP that is, these channels are in scope. This is my working media budget. Tell me how you're going to run this. And you're looking at CPMs and click through rates. That's what you're going to get. They're going to spend the budget as most effectively as they can. If you say, this is what I'm trying to achieve, to your point, profitable growth, or I have to report to my board on how my LTV has increased.
or my CAC ratio, then as an agency, we bring the right expertise together to do that. We're not locked in on a search director and a social person. You're actually putting the right people around the problem. And that's where you do need, you need a different kind of client trust, you know, shared outcome engagement, because that means that, okay, this month we've got a creative problem. Let's pull in our creative team on it.
speaker-0 (31:15.894)
And it it shows that there's trust, right? Because otherwise you're just you're in this race to like mass reach problem, right? We're gonna drive down the CPM rates, we're gonna drive up mass reach, we're gonna spend all the budget. But if you've got a trust real trusted relationship with your brand, right, it's consultative. And you can say, hey, we're gonna we're doubling the CPM rate, Mr. Procurement, Mrs. Procurement, right? For the following reasons. Because we're gonna be in premium inventory of this type that's gonna have a greater likelihood to
speaker-2 (31:40.022)
We're gonna refuse to buy media on just
speaker-0 (31:41.644)
You're gonna you're gonna hit your cohorts in a way that is much more impactful, much more time spent, dot dot dot, right? That's all going to drive outcomes, right? It's it's almost like you've got to you have to convince them and show them that mass reach in and of itself is not a strategy, right? And that that a race to the bottom
speaker-2 (31:58.078)
Maybe for some brands it is, but like for I'd argue for most it's probably not. Yeah. But I guess the question though about if you're looking at like the the goals of the brand, if if if they're just saying here's our money, here's what we want, isn't that kind of going back to the old trading desk model where it's just here here's our budget, go get us a bunch of orders, and then I would and I I think that model invited a lot of the the old fraud, right? Yeah. Like like
speaker-1 (32:21.6)
But but I think that it's e you know, I think that like if you can if you can avoid the fraud, right? Like e at the end of the day it might ultimately be the right model, right? Like, you know, we've we've seen some of the the kind of notions of like demand led coming back into vogue in the industry and and you know I do I do think that there's a time and a place for that. There's you know a lot of brands that it's like, you know, if the unit economics work then I want to push as much volume as I can until you know until I hit that break.
Even point, right? It's you know, the there are moments that you know, I think a lot of times, you know, we as agency people, like, you know, we love all the flashy stuff, we love the creative stuff, right? But it's like sometimes at the end of the day, again, it's like, you know, how do you make your CMO and your CFO best friends? It's like that's kind of a different conversation a lot of times, that's more analytical, more finance led, but that is like really truly ties into the outcomes.
Yeah.
speaker-0 (33:22.006)
Yeah, w which shows you the the CMOs have to be able to make like an analytical and data based data driven argument back to their their CFOs to show that for example
speaker-2 (33:29.824)
And CEOs.
speaker-0 (33:30.646)
The impact of Halo effect in brand versus performance and in certain large organizations, and I'd love if you guys have clients like this, but certainly financial services, insurance, you know, you got a lot of like top of funnel, you know, big brand like linear television, CTV buys, and then you have a lot of direct response, direct mail, and there's very little connecting the both the data that's being shared from customer exposed to this to customer converting here with direct mail piece, or the teams. The teams often have independent, they're independent teams.
With independent budgets, right, and independent incentives, right? And I mean, do you see this on the client? You've been on the client side for so long. Like, how do we actually combine these two to show that there, because there's a mathematical argument that your brands spend from a Halo effect? I mean, PNGs learned this when they spent a lot of money on Facebook and took away from big branding. I think Nike learned this, right? that you take money away from here, it's gonna impact here. And in fact, the money spent here at the top of the funnel with a better balance between top and lower funnel is gonna have a greater ROI.
How do you prove that out, right, to a client to convince them?
speaker-3 (34:34.444)
I think it comes back to where you said AI is not a strategy, right? To flip back to that. We like to talk about this and the shiny thing in our industry, but what you're hitting on is just fundamental how marketing works. Just how people are exposed to saliency of a brand, how you build market share, physical availability, mental availability, all of these things. And our industry, we've pulled things apart. We've broken things.
took creative away from media. We've pulled the funnel apart and we've said there is no funnel and now it's linear. The funnel's circular, there's a flywheel. We love to do What else would we talk about? there is evidence and there's proof. You mentioned Nike and what was the other one you mentioned there of P &G that found this out. There is data, there is marketing effectiveness data from
speaker-2 (35:15.182)
Right.
speaker-3 (35:32.258)
The last 50 years, which proves our case over case, industry over industry, market over market, that a balanced budget of building brand and building accessibility for customers and more seamless and less frictionless way to purchase is a way to grow market share.
speaker-0 (35:49.154)
Yeah, yeah, it's sort of it's sort of the share of voice equals market share argument.
speaker-3 (35:53.678)
Exactly.
speaker-2 (35:54.69)
How can CMOs explain this better to to Brett to your point to the CFOs and to the other like C suite roles? I mean, when you look at CMO tenure, it's now, I mean, it's really low, it's like three years or something like that. Maybe it's even l less than that, maybe it's two and a half years. And many big organizations have r removed the CMO role entirely. They might have a VP marketing and a head of media. So though it's not like they don't do marketing anymore, but I think they've got like there's a trend where many big multinationals have gotten rid of the CMO.
CMO or devalued it certainly. And the CMO might not even be in the in like the executive suite anymore in some ways. So thoughts on like how can that message, how can C Moby help with that message? And like, do you think that's fair? What's happened to the CMO? Like, and you know, the reasons for that could be like, in my opinion, inability to understand like actually like is media driving outcomes. I think a lot of the
opacity of media supply chain and like lack of dollars going to working media has contributed to it. I think the like lack of understanding of the total cost of ownership of the Martech and ad tech stack and a tech tax. I think all of these factors have made the CMOs look like they don't know what they're doing to the other to the other to the other the rest of the thoughts on I know that's a lot I just right there. So like thoughts on that. Is that fair? What's happening and then like w part two would be how can agencies help CMOs tell a better story.
speaker-3 (37:20.204)
Yeah, I do think it's true and it's sad. The role of a CMO in a business is a really, really important one. And today the combination of CMO, CTO or CIO and CFO has to be stronger than ever. Everyone always looks at marketing and goes, cut costs and make things more efficient and make things work harder. It's everything right now and technology is so much involved in that and so valuable to it that it is critical. I think we're suffering from
we've been trained over the last decade, 15 years of short term output, what search has done, what social has done, the immediacy of human reaction to these things, programmatic, that you can do something quite quickly and change a budget mix in digital, to use that word, and have an impact relatively quickly. Now, all you do there is you just take a short term view. And unless you maintain a long term approach to build a brand,
eventually it dives off a cliff.
speaker-0 (38:20.888)
Yeah.
speaker-3 (38:25.216)
Yeah.
We've all seen that char.
speaker-1 (38:30.478)
the easy answer. It's like we can prove it with my god meta did so great or like search was so great right like y yeah but it like those things happen in combination with all of the brand awareness. Yeah and it's like it's it's that whole build. And so you know it I think like to the question of like what can brands do. I think you know there's a big argument like one of the things that that I think as agencies that we can really help with is
Is that consultative layer around data and around measurement and how do you do, you know, not just one thing, right? But it's like a whole program that looks at like MMM and like, you know, and and causal impact analysis and different type of attribution models and you know really helping them think about it holistically in a way where it's like, yeah, bring all of your data together, you know, structure it so that you can actually measure it, so that you can have
have proof and evidence of you know the the effectiveness of your marketing. but then that also you know gives you a really solid foundation as a place to start your AI strategy from, right? It's like if you don't have good data.
speaker-0 (39:47.438)
Your electricity.
So I mean would you love this is kind of it's like the attribution fallacy and I think I think part of with the greeting your own homework walled garden approach where it's like yeah within this ecosystem, within the Amazon ecosystem or the Google ecosystem, you know, we can attribute at X percent confidence, you know, all of your but there's there's oftentimes claims to c cr like cre they're you're crediting either channels or techniques that really didn't deserve full credit because there's multiple exogenous factors that are impacting.
speaker-2 (40:17.622)
Well there's the famous Uber meta example where they actually they they reduced their meta spend to zero out of from like two hundred and fifty million and it had zero incre inc incremental impact. It wasn't driving a single incremental download actually.
speaker-0 (40:30.348)
Yeah, an incremental contribution. I mean incrementality is like that term ad nauseum we've heard a lot, but an incremental contribution, meaning are you are you adding customers to our you know to to our our bottom line that wouldn't have come otherwise, right? Either through like natural propensity, seasonality, whatever it might be, economic shocks, media exposure across multiple channels, like how do you actually causal impact analysis to your point helps to get to the bottom of like what really is driving the impact and then how can I quickly pivot?
'Cause the challenge I think with MMM traditionally was that the pivots were so s there were no pivots. It was an annual six month planning process, right? Pulling in all this data, then you do an output and then you basically invest for the next year, right? I mean, how have you seen brands in terms of their kind of their sophistication in terms of making these moves quicker to be more responsive, more proactive?
speaker-2 (41:04.91)
Yeah.
speaker-3 (41:22.03)
I mean, some, some it's still still a journey, I think, and that that is the role of us as an agency. That's where we have to be accountable. I love doing that, by the way. I mean, just, you know, if you're having a debate with a client and they, you know, well, we can't just turn it off. Just say the amount of times we've done that and it shows it just there's no better way to make the case. Working and if it's working or if it's not, you know, immediately, will know immediately. And I think that is having
It's less about the education and the pivot. know, a senior marketer trying to turn the whole organization, as we said, to get in a different direction. But it's, think, having the bravery almost to ask your agency partner the right question and say, this is actually what I need to prove out. How do I show that what I'm doing is right, wrong? What should be my next move? If you can give your agency
kind of the carte blanche to help you with a question like that versus here's my $10 million budget, go spend it. Then you're just gonna get to such a different place and improve value to your stakeholders internally.
speaker-2 (42:32.91)
So I have a question about like how the agency model is responding to the differences in, let's say, user experience. And I and what I mean by that is
You look at like as all of these, let's say, solutions go headless, like Hell Salesforce went headless, right? Yeah, meaning like to those who in the audience who don't know, like they've basically taken built MCP servers so you can connect all of like what is SaaS? It's databases, workflows, and like a UX UI, right? Where you can control all the things underneath. As AI, as agents begin to do things instead of people, in theory, like agents can interact directly with these databases and these workflows, not actually have to deal with the complex UX, which I think would be great.
'Cause like the whole you th I think UX's UI got so convoluted with like fifteen, twenty years of SaaS like overc over complicating it. So I think on one hand it's great, but given that at some point all the DSPs go headless, all the like people won't be in platforms any well anymore, it'll be agents doing it.
On one hand, you could argue this is an incredible opportunity for agencies. Agencies can be we can we can create agents for you brands and we can take these agents and we can go across multiple platforms to accomplish one thing for you to drive towards outcomes. But you could also, the counterargument to that is, well maybe brands want their own age agents. Once brands have their own agents, what are agencies doing? So thoughts on the positioning of the agency and like who has the advantage in this.
speaker-3 (43:54.88)
I'll take it. That was a tough one. Yeah, that was a good one. Now we're getting into it. I think it's both. think the role, honestly, of an agent, I think that will happen 100%. And I think we should really advocate for brands having their own agents and doing that themselves. Honestly, nobody understands their brand better than they do and agencies ever will. But I think that one of the richest areas that's not going away is, you
speaker-0 (43:56.354)
Yeah.
speaker-3 (44:23.622)
brands data and just like brands themselves, right? They live in the minds of consumers. The way you think about P &G versus the way I think about it is very different. The way you interact with it will be to everybody and then everybody in the city, right? There's so much rich understanding that, you know, exists about it that it's not just, that's not the end of the line. Having agents to then go and buy media and access these things in different ways is
is an amazing advancement and will happen and will be incredible to see what impact that has. But the role of an agency to really unlock what is distinct and unique about a brand and get into market in interesting ways and help them understand consumers better. How do you make this more meaningful? How do you understand what your competitors are doing and ultimately help them grow, think is not, it's just going to morph into a different way of operating.
speaker-0 (45:19.638)
I mean do you think do you think brands will run in I mean t with that sort of agent doing the buying for you and whether or not agencies fit into that sort of world, right? I think you kind of run into the problem of in housing. Brands often try to have have gone through this path a couple.
speaker-2 (45:33.55)
There's been several cycles, even in my career, right?
speaker-0 (45:36.334)
You're gonna in-house media buying and they're like, well that's a disaster. That didn't work, right? Right. You realize the complexity of the ecosystem, the number of channels that you've got to manage, and then suddenly the brand is bringing on these massive FTE headcounts to try to support they do the same thing with data science and analytics teams.
speaker-2 (45:52.908)
Right. But maybe they don't need to bring on the FTE headcount anymore, w when you go to a agentic model, right? I mean maybe they don't need to.
speaker-1 (45:59.661)
Yeah.
Yeah I mean I think I think there's like like there's there's a logical argument for that, right? Like, you know, I mean we've already seen that that like really all of the platforms can you you can interact with them in a headless way now. Like you know, we've like like I would I would argue that that's already you know, that's today right now. Yeah. and and I think that, you know, and and w we've been working in a much like a similar way. It's like, you know, very heavily leaned into MCP servers, like how do you
give all of this control and ability to access two agents, but in a really like highly governed and regulated way. And so, you know, I mean I think that there's there's a moment that like intellectually you could say, okay, you know, like a Gemini agent is as knowledgeable about like media planning as, you know, like any of the media planners that that I've worked with in my career, right? Like on paper, absolutely. Like the it it performs exceptionally well.
But there's also a moment about like how do you bring everything together and really coordinate across the entire problem, which is where you still have you know that that need for human agency in the middle, which is you know, understanding of the brand, understanding about like what truly works, what doesn't, you know, having taste, having, you know, all of those things that that still are the very human element of what we do.
And and I think that you know yes, are there you know, is there a different role for agencies to play in this world where it's not so much about like putting FTEs to like media buy through platforms? Like absolutely, it's different. But it you know, one of the things that that I think that most people in our industry identify with is you know, I love this industry because it's one of the fastest moving, most challenging industries that I've ever seen. Like for me.
speaker-1 (48:00.992)
it's an you know it's it's an intellectual challenge to do advertising because it it's always different right it's like you know every year it's the year of the something new right and and so it's like like I actually love that and so you know it's really that kind of like adaptability of how do we change what we're doing to meet the moment of what's the technology look like, what's the media landscape look like what's consumer behavior doing you know like the change is actually really you know it's an opportunity
Yeah. And and it there was a quote yesterday that that I really identified with is you know, the people that lean into AI and agent and agentic behaviors early will get a sustained advantage, right? Like, you know, there is a notion of like wait too long.
speaker-2 (48:53.716)
Not just companies, but people.
speaker-0 (48:55.342)
Continuous learning. But but back to that argument about brands. But I do think that brands have, you know, their competency, their kind of core function and purpose in life is to sell, build, package, and sell products or services, right? So they've got to be the master of that domain. For them to hire Adam Woods, the chief product officer, to build a stack, to have engineering teams in multiple places to build the complete, like a like a you know a complete system from data infrastructure to MCP layers.
To you know, the harness for your agentic intelligence stuff. Like it's a lot to ask for for brands to in-house a lot of this. So I would, I mean, if I had to prognosticate in terms of like you guys seem to be moving in the right direction from an agency perspective, agencies have got to play a more higher value value consultative role, right? Which means they need to automate certain things that took a lot of production time, that weren't intelligence, that weren't insights, right? And that and that the agencies can play a role both with tech and
services that that brands probably aren't gonna want to do entirely themselves in-house.
speaker-2 (50:00.546)
Well I think there's been a tr you know, th you mentioned before the trend of in housing every few years there's always gonna be a bunch of in-housing and then it kinda works, kinda doesn't, and then it's spun back out. So I think it's probably gonna be easier to in-house more things with AI. But I do I'm gonna answer my own question from before. I actually do think the advantage goes to the agencies, right? Because the agencies will be able to build better tech in theory, right? that than brands will because it just I mean you look at if you compare an in-house agency to an agency agency.
And in house in the in house agency is gonna have a harder time at attracting and retaining the highest skilled people. Yeah. Even if you need fewer people, they're gonna probably even be have to have more skills. So I think advantage goes to agencies over in house agencies. I also think when contracting, I mean in housing sounds great until you realize you have to pay Google and Meta and TikTok. Not thirty every month. If not, they're gonna shut you off on on day thirty one. Yeah. I mean every I mean clients how how many clients pay net thirty?
speaker-0 (50:53.398)
Like the order management and billing and all of that incredibly complex.
speaker-2 (50:59.638)
Zero, right? They paid net sixty, net ninety, like I mean, you know, some of the big telcos will say net ninety to rise, net one ten, and guess guess what? Like like you would have been shut off like ninety days ago, right? Yeah. Yeah. Eighty days ago.
speaker-0 (51:10.862)
And the agencies are taking on that some of that.
speaker-3 (51:12.718)
I was going say how many clients know that you have to pay at 30.
speaker-0 (51:15.338)
It's a management, you know, agency as financial services.
speaker-2 (51:15.896)
Yeah, yeah.
So you're paying those platforms every month, right? And you maybe you have a line of credit, maybe you're not, maybe you're you you're but you're maybe managing the float of that, but a reality is a big part of the agency is representing these brands and buying all this media for them and interacting with the platforms and and making so they don't ever have to worry about that. So I I think that's the easily overlooked aspect of in-housing that's a freaking guy. So think about it. Yeah. Yeah. I don't think agencies I think agencies will change, but I don't think there's certainly
speaker-0 (51:38.134)
I think we're I think we're making the argument for you.
speaker-0 (51:48.384)
So one thing I want to cover with you, Adam, because know you love this topic, and we've talked about it like offline a few times, is this notion we talked about data sovereignty, and like I'm not to get too geeky for the audience, but the idea that, yes, a brand controls its own first party data, all the data that it collects through media performance insights and all that stuff that comes into their own cloud ecosystem and they manage and optimize everything they're doing around that, right? What about a gentic? What about a gentic sovereignty and this sort of context layer that brands are building? So as an agency or even as a tech.
You might be building this incredible model that does auto optimization, creative media by whatever on behalf of a client, even in their own environment. But if that model's the predictive capacity that model's building, and it's being trained on that, is then being used to bring to a client, right, to transfer some of those learnings, right? Like, which is part of just how the model learns and gets better at what it does. Is does that put sort of a general?
speaker-1 (52:51.064)
Yeah, and and and I think that it's you know, it it's it's things that like we've all known kind of like as industry patterns for a long time, even even before AI and agentics, right? It's it's this notion of like what what is your your true competitive advantage, right? What are the things that you can do that others can't that give you a right to outperform your segment? Yeah. And and then and then being protective of those things, right? And and and a lot of times
like you know even protective around like your risk management so you know when we talk about data sovereignty you know a lot of that thinking comes from having worked in highly regulated industries where you know you go to the CIO of like a large bank and it's like their data is in their environment for a very good reason right yeah and they're not letting go of it highly sensitive highly
speaker-2 (53:45.614)
There's no.
speaker-1 (53:46.798)
Like it's not going anywhere. And and and I always found that model to be a better operating model because it's like, you know, yes, you should want to protect your first party data. You should put it under your security parameter. And then we think about that in this in this world of AI and agentics. And and it actually, you know, it makes the stakes even higher. Right? Now you've got a lot of like non-deterministic systems that have access into your data that can
Can do things that are highly unexpected. Yeah. Headless across your entire sources with IO and you're thinking about this, like part of your core job responsibility is mitigating these risks. And how do you mitigate those risks if you've got like every agent under the sun connecting in with no regular, you know, like
speaker-0 (54:19.594)
And can go across headless across your data from multiple in your ecosystem.
speaker-1 (54:41.994)
Yeah, it's i it's absurd. Like it's not risks that you would ever take. Yeah. And and then you think about it also from kind of that like competitive edge or like intellectual property. And it's like you need to, you know, to Amy's point earlier, own your intelligence, right? Don't rent it, don't let somebody else hold it hostage, right? If you're gonna invest in this, which is which is how you
speaker-0 (55:04.386)
Carefully like like literally do in your contractual arrangements with vendors. Like this has got to be down to the line of like no, you're not gonna use this to train my competitor. I'm Unilever, you're not gonna use PNG, right? It's not just the data that they're using, but it's it but it's everything that you've learned, that institutional knowledge that's being put into your your outred.
speaker-2 (55:15.68)
Yeah, you certainly don't want do that.
speaker-1 (55:24.654)
Yeah.
speaker-3 (55:26.094)
things.
speaker-2 (55:26.296)
Yeah. I theory these LLMs when you're using them they're not supposed to prompt itself, they're not supposed to actually be storing that, right? But the th thing is like if you give it a new idea that it didn't have before, like it will have that idea still, right? So it's so like you're
speaker-0 (55:40.352)
Your idea as a brand, right? So I think there's certainly ways that you should protect
speaker-2 (55:44.11)
So imagine you're like a life sciences company or like there's no way you'd want to share like your RD with that, right? Because your competitors in theory you'd want to have that locked down to your poor, there's no way you'd want Or financial services is or or if you're a hospital, there's no way you're gonna want you legally you can't even share patient data, but there's no way you're ever gonna want that anywhere near one of these frontier models.
speaker-3 (56:01.464)
Well, that's a, sorry, that's like fundamental change of agency, I think. This idea that we you know, we build a silver bullet approach over here. We're to spend billions of dollars on data and we're going to build a massive tech stack and then clients are just going to sign up and use this thing and pray a premium for it. No, like that, that completely goes against that whole case. What you, what we need to do is help them.
build the model from our experience working across category and industry and then guide their people to build it within their environments.
speaker-0 (56:33.09)
Yeah.
speaker-2 (56:33.496)
That's interesting because there are really two models you think about it for like what what I see agencies and hold coast building. One is saying we're gonna build this like take audience data from across all of them. Maybe maybe you can argue it's like put in clean rooms or it's pseudonymized, but we're gonna build this infrastructure and have this data we're we're gonna use to do things better and more efficiently, or you could say we're just gonna bring like the
the architecture for you and you bring your own and you're never gonna ship it here. We're gonna it's gonna be a zero copy and w and you're gonna take this and and we're gonna use this to execute for you. I think these are two very different models.
speaker-0 (57:06.094)
Yeah. And I th I think to your point that it seems like the model's moving towards that no data movement. Right. You could say the c clean rooms are finally seeing their day in the in the sunshine, right? Like they've been an empty mousetrap for a long time. They're like, well, what data are we putting into the clean room? Like do we need data partnerships? Right, but they're they're finally seeing where you can do a lot of this.
speaker-2 (57:24.878)
Yeah, did it didn't it didn't survive as an indiv independent category of tech, but it they certainly are coming I think they're finally becoming like a thing people use.
speaker-0 (57:33.324)
Yeah, and without surfacing any of the PI or the sensitive information.
speaker-1 (57:36.538)
It's it's a useful ingredient, right? It's you know, i it's a way that like, yes, you know, if I want to bring first and third party data together and I don't want to share my first party data, right? Like like it it it is it is a good capability. But at the end of the day, I think that it you know, it didn't solve enough, right? And and that's where, you know, it's kind of like, yeah, when you can use a clean room to then actually go really wide and look at like, you know, how am I looking at incrementality?
Exactly, yeah.
speaker-0 (58:10.52)
New audiences that are trying to audience overlap is a common case case use case, right, with partners.
speaker-2 (58:17.138)
I like what you said, it's capability. I clean room is not really a category, but it's if you want to be able to zero copy, collaborate multiple data sets, you're basically describing a clean room, but you don't necessarily need like an off-the-shelf clean room to do it. You just need a place you can do that where you're that has certain like r controls on it, so you can't see like row level data, you can't export in certain ways where you actually have zero data movement. That's really all you're trying to do, right? Yeah. So
speaker-0 (58:42.956)
Yeah, and it facilitates the partnership with publishers that have a lot of audience and consumer and cohort based data, right? That they don't want to give away because that's sort of the golden goose, right? And so how do they actually share their data where you can do analytics and insights, but you know, without without them feeling like their their data's just gonna be ingested by another platform and used by a bunch of brands. Right. So I wanted to go to to to Amy because you and I talked about this offline at one point, about the level like we're we're having this conversation about clean rooms and about a jitticate. I mean a
speaker-1 (59:01.13)
Yeah. Yeah.
speaker-0 (59:12.96)
about AI sovereignty and data you know and a lot of these topics are kind of esoteric. I mean even though they're based firmly in reality. We're not making this shit up. This isn't science fiction. This is actually right now, here and now.
speaker-2 (59:24.214)
Maybe we're making some of it up.
speaker-0 (59:25.62)
But Amy, you've talked about how just the translation layer between all this talk that we have in the industry, right? When you actually go into a like a client and and you know, depending on the seniority of who you're dealing with and the function that they they live within, let's say within the marketing org, and how there's a lot of education and hand holding to bring them up to kind of because again, agencies come to the table with a lot of cross industry, cross category knowledge and insight and wisdom and years.
of experience. like how do you manage that? Like how do you actually bring clients along f in the journey? Is that is that like what keeps you up at night?
speaker-3 (01:00:05.274)
Yeah, I mean we started talking about kind of transformation in people. think it is, we've got to, we're trying to distill in our people an idea of just, it's our job as the agency to do that. If a client isn't on the right journey or they're not really trusting us or we've not got the shared outcome, it's because we haven't done the job well enough to educate them on that, right? That's the first thing. And then I think in, yeah, I mean the whole agency model is changing, so it's no wonder clients are...
know, confused about it and where we can play, but it really is, it's in the small things, honestly. I mean, the amount of RFPs or pitches or client conversations we've started where it's all AI, all agentic, all the dream state in reality, it ends up being, you know, so, so, so condensed. And the starting point is often something so simple with a client of just looking at their creative and making everything, all of their assets globally consistent.
make everything look like Oz. First task, four to six X every media metric they've ever measured. Just by doing that. Just by doing that. It's the simplest things. So it really is about finding this balance between dream state and all of the questions was the lowest hanging fruit. Yeah, the lowest. But how many brands do you see that don't get that right? We talked about the break between brand and performance. We talked about some of these
really simple things that have been around a long time. It's not like the conversation in our industry, you know, we always have something new and shiny thing to talk about. We don't really solve the thing from the prior year, you know, we just move on. Yeah, exactly. And so.
We have ourselves. Yeah. Yeah. So it's going back, finding that low hanging fruit and proving the value there and then using that to build momentum.
speaker-2 (01:01:49.602)
We figured out what the last
speaker-0 (01:01:57.538)
Yeah, and and like we talk we've talked about this a lot, this idea of like the the the board mandate at the brand let's say at a brand, right? That's like we are going to become AI first. Again, AI is not a strategy, right? And so the board is a little bit disconnected from what's happening on the day-to-day basis, right? We to we've talked to this with creatives, with media people, and then at the day-to-day level, they're like, We actually already have a lot of automation built into the things that we're doing, whether or not it's what you're calling AI, like the way you understand it from what's happening in the media, LLMs, the frontier models, right? Is almost
Almost i irrelevant, right? And so it's so it's you know AI can be implemented in a lot of ways that aren't exactly new, right? There it's machine learning one on one of like things that have been automated to help us controllably do things better.
speaker-2 (01:02:42.306)
But it's funny you mentioned that too. So like sometimes and I guess to kind of pile on the point you just made a second ago, Amy, about companies, they always want to do s like the this big elaborate, expensive thing where sometimes it's a a kind of a simple solution. I'll give you an example. I was working with this, I think we might have talked about this one at one of the pods one time, Brett. I was working with this regional bank and they had this app they had deployed.
And they were they told me come in, bring your whole UX team. We think this app is garbage. No one's using it. The performance is terrible. And so we did this entire audit of it. And after the we concluded this audit, you know, it took like four weeks or whatever, hundreds of thousands of dollars. I mean my conclusion was the thing works perfectly fine. It actually d looks pretty good, but your communication on this is completely wrong. You're not giving you're not communicating what it actually does, the value proposition. They wanted it to be this like app you could use to like
Manage your investment in your portfolio. Meanwhile, it's regional bank. People are gonna go with like Schwab or E-Tray. They're not gonna do that with you. But it was this great concierge app where it could like you you wanna you want to manage your car loan, you want to manage your mortgage, you wanna look at your bank account. It could route you to these things perfectly, which would have taken multiple calls or going on their website. It were it was great. I said just communicate on communicate that. People are gonna use it because it's actually really good. Like whoever built it like did a pretty good job. I could you maybe prove it a little bit sure.
It wasn't you didn't have to redo this whole thing. Of course they didn't listen. They hired someone else to get to get the answer they wanted and then I I still think it's probably not used, but sometimes it's just sometimes it's just a simple answer, right?
speaker-3 (01:04:17.358)
Yeah, I love that. I when I kind of started out in media and strategy, was on a lot of government campaigns and behavior change. And I still to this day find it odd that you get a brief that's, okay, this is our KPI and this is the budget. And how amazing that the media plan is exactly that dollar amount to achieve that KPI. that. It's like, yeah, if you're doing, you really, that approach of, this is what we need to do, well.
sometimes it's not a media thing to get some, know, and especially dealing in government, getting people to stop speeding on the roads or look out for cyclists. That's not a TV campaign. That's a little discreet button that you put behind the opening the door and like everything, you know, at our disposal is media, is communication, it's creative. That way of thinking about it of what is, no, this is fine. Like you don't, we don't need to invest and not having the right answer all the time. That's where the most interesting work comes in.
speaker-2 (01:05:11.086)
Yeah.
speaker-0 (01:05:15.894)
And from a client management perspective, it sort of seems like if if the agency would have the sort of ability to sort of walk away from the short termism of like we're gonna spend every dollar you have and come back with a strategic recommendation that may not spend the whole budget or maybe reallocate the budget into different areas. It's kind of a bold thing to do as opposed to just being the yes man and saying yes client.
speaker-1 (01:05:41.644)
But it but it's the right thing to do. And and we've seen, I mean, the there's been RFPs that we've you know that we've looked at or participated in and you know it's like it's like the the the brand or the client is like, you know, no I want like I want to go down this lane, right? It's like show me your rate card, like it's gotta be billable hours. And you know, the there's sometimes when we've you know, we've walked away from some of them and just been like, you know what, like this the no
matter what happens, this isn't gonna be the right kind of business that we want. You know, it's not gonna be that partnership that we want. and then there's also been, you know, there's been a few cases where it's like, you know, take their ask and come in as the complete wild card. Right? Like here's a different model. If like I know you're not gonna be able to compare Yeah, like here's here's what we would do for you. And you know
speaker-2 (01:06:34.742)
Yes for A we're proposing B.
speaker-1 (01:06:41.418)
Yeah.
speaker-3 (01:06:42.57)
This is the RFP
speaker-0 (01:06:44.622)
Yeah, yeah, you don't sound like everybody else, right? Right.
speaker-2 (01:06:47.086)
See, I'd rather do it that way. I'd rather go in and say, you know what, like here's a better way to do this. Like like we've done this before, even if we haven't, like we think there's a better way to do this, just unlike what you're proposing. Because I we could try to propose the way you want, but it's not the right way to do it. As a as your agency, like we need to tell you this in fact.
speaker-0 (01:07:05.89)
Yeah. What do you guys think about outcome-based pricing? Because I know outcomes-based agency people are going to start to think, Ria, I know Rio's got some strong perspectives and probably has had clients push towards, hey, why don't we do outcomes-based pricing? And I think the general comp well, let me ask you that, and then I'll say what we've what we've seen in our own worlds, right? Like like when when a client says we want to do outcomes-based pricing, what are they really saying? And and does it ever work in favor of your interest?
speaker-3 (01:07:29.006)
It can, it can in a shared and common goals around it. But I think where we'd like to have that model in play, we had to like to have an element of that in, but agreeing on what that outcome is, is always the first hurdle. And I think unless you are starting off a relationship like we're talking about with the right thing to do, with a trusted shared outcome, if it's this, you're my vendor.
then no, because there's going to be things that are completely out of our control and it's just going to breed.
speaker-0 (01:08:02.188)
got all the leverage and the the negotiation power, right? And so they're gonna put all the risk on your shoulders and try to de-risk as much as possible on their end and you end up just, you know, if it fails, we're screwed.
speaker-2 (01:08:13.602)
This this is what I've seen. So I've seen like I've had a bunch of clients over the years ask me, like, can you give us a an outcome or performance or value base? We'd call them for a while. Contract is funny. And like
I don't think I've ever had a client actually say yes to a real value-based contract, meaning we'll do it for at cost, like for half of what we would, or for a certain production, but there has to be an upside for us. We'll take risk, right? We'll take risk, we'll make no money, or maybe even lose a little money. But if we overperform, you're gonna overpay us, right? I've never had a client, like when you put the actual contract in front of them and they see, crap, I may actually pay 1.25 times this.
that that would maybe overperformed. The potential of us overpaying, scared the crap out of them so much, they would never actually say I'd say, okay, actually just give us the normal contract. Or or they say, do it for us for less. What they really want is a discount, right? Which is a different conversation, right? They say, okay, if you want a discount,
Fine, maybe maybe we can figure out a way to get a discount in like for maybe longer. There are things you can give them, it's like a quid pro quo, right? Maybe to do that. But I've never it's funny, I've never had a c client actually say that. I do know one guy, it's funny, my my old company. There's one guy who did good client to agree to a true value-based contract. And of course we cr you know, we crushed it. I remember that first year. No, no, listen, like it was like the year ended and they said, actually we're not renewing they they were gonna be hit with that big
And he said there's no way you just try to collect.
speaker-1 (01:09:44.79)
Well i and and I think that's where you know you have to be smart about it, right? And it's and and I think that it it it needs to be like an element, but it also needs to be balanced, right? It's like go into it with like, you know, we've covered cost, we've got, you know, some margin that we're making, that like, you know, we all believe that this is a very fair value exchange and then there's you know a a performance component on top that helps you have like shared alignment, right? And and it's you know, I think that's where you win, not on the
like highly speculative like huge upsides 'cause I I think I mean yeah, you run that risk of like non payment or you know like you you overperform your way into churn right
speaker-0 (01:10:26.912)
Overperforming way into church. Well this is like we've we've gone over. We've got another another another podcast book, but this has been great. Thank you for spending so much time, Adam and Amy. coming down from from Woodstock, coming over from Denver. Well you're at the
speaker-2 (01:10:28.59)
Yeah, yeah.
speaker-2 (01:10:44.24)
For ad week properly, right? Yeah, yeah, yeah.
speaker-0 (01:10:46.408)
Adam's on the road like 180 days a year now, right? You're like James Brown in his bang, right? hardest working man in show business over here. But again, thanks Incubita, thank you guys for joining us. it was a really great conversation. It these are these are topics that are sort of you know at the bleeding edge of all the conversations we're having, I think, in terms of what's really happening and how do we make sense of it, and that's kind of why we exist, to try to learn from you, from you guys, and also to make try to make sense of of some of the the noise that we hear in the in the ecosystem. So thanks
everybody for joining us. Thanks Adam and Amy and we'll see you next time.
Comments