Welcome to the MadTech Podcast from ExchangeWire, bringing you the latest news and views from the world of ad tech, media, and marketing.

Hello and welcome to this week's MadTech podcast.

I'm excited about this week.

It is a very special mailbag edition where we have three questions from our audience that I'm going to put to two of our senior team.

Welcoming CEO Rachel Smith.

Thank you very much, John.

Delighted to be on the mailbag special.

I know I'm pleased to have your insights for these beautiful listener questions.

Welcoming also Lindsay Rowntree, COO.

Good, are you ready to answer some mailbag questions?

I definitely am.

Amazing, well we've got a really good selection across the ad tech industry.

I'm going to kick us off with a question that has come in from Barney Worfolk-Smith of DAIVID.

um He has asked, so Adobe expects Gen AI content to grow by about five times.

But 70% of Gen AI ads already generate below average positive emotions and are 13% less trusted than the norm from our research at DAIVID.

Is the industry sleepwalking into a creative quality crisis?

Ugh, crikey, I've mangled that.

But it's pretty hard to say.

uh

Is the industry sleepwalking into a creative quality crisis and who's responsible for catching it?

Thank you for that alliteration Barney.

Rachel, what do you reckon?

How is AI impacting creativity?

Who's responsible for it?

So um I think that I want to start by uh sharing with you a little anecdote from a friend of mine in Sydney who is a computer animator.

Um, because he and I got into a conversation in January about the role that AI was playing in the creative process in his work.

So he works for, I mean, he's worked in uh film studios and TV production studios doing computer animation.

Uh advertising agencies.

And now he works for a special effects company that does special effects for all of those sorts of industries.

So his primary role is inserting pieces of creative video content in particular parts within some kind of video.

And so we got into a conversation about the role that AI was playing in his day-to-day job, and he said that they were actively using AI tools, including him, to create.

portions of videos that then get overlaid with special effects, sometimes animated via computer, obviously, and and then teaching AI models how to then perform those tasks

without a human performing them.

And he said that it was both exciting and terrifying because he was delighted with the amount of time that that was giving him back and the amount of time that he would

ordinarily pour into going over someone else's work.

was now being significantly reduced.

And he also wasn't having to hire and pay um designers to necessarily complete all of those tasks.

The other thing he said was his early days experimenting with some AI tools that were available on the market at the time.

The other thing he said was that they were extremely good quality.

So that was the part that was terrifying.

He was thinking, not only am I now no longer employing a designer to do this task, I'm really impressed with the output.

But then we got into conversation about how we even get to that output.

And I was insistent that the only reason, well not the only reason, one of the main reasons why the output was so good is because he was prompting, he was prompting the AI

tool to refine, to he was giving it the initial brief.

He was getting whatever and he kept saying that whatever he was getting back in the initial stages was really good quality.

So that's because you were giving it an extensive brief because you're an expert and because you know what you're looking for.

So my feeling was one of, you know, a mixture of these AI tools that we're all coming across in our day-to-day lives for sure are going to enhance many things about the way

that we work and producing creative assets, I don't think is any different, but that actually the real skill will be in some form of human interaction or human instruction to

make it truly distinctive and different, non-generic.

and something that stands out of the crowd.

And when I was thinking about my mate Jez's example, that's precisely what he's paid for.

So My mate Jezza.

My mate.

Australian, can't have a full name.

Can't have a full name.

No.

Have a really he does actually have a normal name, but nobody so um but he you know that is actually what he's paid for is to sprinkle that kind of human magic over what was

already a digital industry, right?

So you know, and even when he graduated at university, he did like an industrial design degree and went straight into computer animation.

And computer animation at the time was talked about as this thing that was overtaking manual animation and really changing the way that.

You know, films were made and constructed, like he works on the original Happy Feet film, things like that that were of their time, completely different to anything we'd seen

before.

So there's a part of me in all of this, in Barney's question, that thinks there is every reason to believe that this can be an enhancement on creative production rather than just

sleepwalking into generic nonsense.

But it is going to require human intervention to make sure that.

The sleepwalking scenario doesn't happen.

And if we think about this and its use and application in an advertising context, yet again, you know, Jez is like trying to make great cool videos for film and television and

sometimes advertising, but Barney works for an ad tech platform that's trying to improve the quality of advertising, and that's sort of our day-to-day business.

There's no reason why we can't be using those AI tools to improve quality rather than to reduce originality and to create something generic.

But we have to really be switched on.

We have to have our eyes open.

Otherwise, you know, we might, we might end up in the sleepwalking scenario.

The thing that I the the the bit that makes me hopeful is that our entire industry has been moving away from, if you like, a generic, uninteresting form of advertising that was,

you know, until this moment fueled by

cookie-based targeting to something that has to be by design much more personalized because it needs to be opted into, it needs to adhere to privacy constraints.

And so therefore the entire industry, I think, is actually, you know, by design or by by accident, moving towards a more intentional form of advertising.

So I'd like to think that inherent in Barney's question is what's going to happen to the quality of advertising.

I'd like to think that there's some natural

pressures in in the industry at the moment that will drive us more towards quality and away from generic nonsense because generic nonsense isn't going to work.

It's not going to produce the results that brands and advertisers and marketers are looking for.

Lindsay.

Yeah, I mean, I love the anecdote because I totally agree.

I think that when you remove the barriers to entry and kind of democratize something, the barriers are normally cost and with cost comes regulation, oversight, everything else.

If you make everybody be able to access something, you're removing the aspect that kind of makes it then, um you remove the checks and balances.

And that's what we are now, we now exist in a world where barriers to entry are being completely removed.

And so within that,

checks and balances have been completely removed and we need to find a way of bringing that back in.

And often cost is the thing that does that.

So if you remove the cost limitations, and I don't mean obviously like AI is expensive and using credits is expensive, but how expensive is that versus a full like 30 second TV spot

shoot that requires a full studio, it requires uh runners, it requires production staff, requires that, whatever it requires like that versus how many credits you need to create

an ad on AI.

And no matter how many people you've got, like what was it, it was the Coca Cola ad from last year where they Coca Cola were trying to come out and say, yeah, but we had a huge

team.

Like there's so many people involved in making this.

It's like, but it's still AI and it still kind of freaks us out.

DAIVID's own research and I've seen Kantar research as well that sort of backs this up that people aren't necessarily against AI ads.

Like AI ads.

not can, will neither break your campaign or kind of make your campaign.

It's somewhere in the middle.

The only reason it would break it is if it's shit.

And that's the problem I think is there's so many shit AI ads.

And again, it's the human touch.

It's the human oversight that I think was going to cause a campaign to be successful or not.

And like, and again, a lot of these different researchers, I think DAIVID was part of that as well, was that AI ads will generally elicit an emotion, but it isn't necessarily a good

emotion.

It can make them angry.

And I think Kantar did an A-B test in Europe with two Coca-Cola Christmas ads.

One was an animal based one, know if you guys saw it, but it was like a rabbit and whatever, like frolicking in a wintry forest.

And the other was the famous one that involves people and everyone preferred the animal one to the people one.

And it's that uncanniness around having people in ads that still freaks people out.

Like, cause you're looking at it, you're going.

something isn't right about this, but I don't know what.

And I keep seeing eToro ads on TV, on streaming platforms, eToro, the of the investment platform.

And there's some weird kind of like, he looks sort of Swiss with like shoulder length blonde hair.

They've designed, they've made him look a bit like a kind of unassuming, you know, Northern European man.

And he's AI, it drives me mad because he's clearly AI and it really annoys me.

And I think part of it is I wish they would disclose it as AI.

If they just said this is AI directed at, at the bottom, I'd be like, okay, fine, but don't make me have to guess.

Don't make me have to try and work out if it is like on Instagram and on the social media platforms, you see these kind of like, either this kind of Amish elder or this kind of

Chinese auntie or, you know, these people kind of saying, if you see a pepper that's this or this, this is a male pepper, it's all absolute bollocks, but it's...

And you're looking at going, are you a real person?

I really feel like you're not a real person.

You're trying to find text around them to see if the text is mangled or their teeth, because their teeth keep moving.

And you to look at all these things and we shouldn't, as consumers, we shouldn't have to be doing that sort of stuff.

So either you declare it or you don't declare it.

if you are going to kind of go down the AI route because it saves you money, you need to spend some money on the checks and balances and the human oversight.

In a world where,

regulation hasn't caught up with how fast AI uh is developing.

I think that's what we need in the meantime.

We need to kind of like be human about it and be normal about it and be like, well, okay, if an AI ad is kind of okay, but people want to be able to know it's AI and people don't

like humans in AI, don't use them.

there's an Instagram, someone I follow on Instagram, I think it's called Sergio Cilli.

He's a director, he directs ads and things.

But it's satirical.

He does these videos where he kind of goes actors versus AI.

So he'll give them a prompt, like he would prompt an actor in an ad and ask them to kind of like just say this.

And it's always an absolute car crash.

And one of them, he's like, right, say this.

And the bloke pulls out a gun and he's like, is that a fucking gun?

And it's like, yeah, and it's really funny.

But the point he's trying to make is like, we do still need people in this world.

And actually at the moment, for the most part, humans will not be replaced by AI.

but AI is definitely becoming a real thing when it comes to advertising.

So how do we make it work?

And there are lots of studies being done.

I think it was 2024, Kantar did a study where it was, I think they do an annual Attitude to Media.

And so this is a couple of years old, but I imagine it's not changed.

The percentage of consumers that liked AI ads.

the percentage of people that run marketing campaigns that liked them, there was a disparity.

The people that are running the campaigns shouldn't...

Exactly.

You like it because it's easy for you to do a job when you think it's the future.

But actually, if your consumers don't like it, you need to have a rethink here as to what's going to work.

And I think that we can certainly make AI effective.

And there's already proof that AI ads are effective, but we need the human touch.

I guess the fear

I said the same thing about 10 times over there in various guises, but you know I'm getting out.

It's a really entertaining journey though, I think.

Thanks very much, Lindsay.

Um I I think I think probably the fear, listening to what you were saying, Lindsay, I think the fear is, and I completely understand where this comes from because I I do think

this is a real concern, that we'll sort of morph to if we accept that there's already some pretty average AI ads out there and there's some ads that you talked about that were, you

know, that people were surveyed on their responses and attitudes towards them in studies.

So there are some AI ads where with people in them that people don't really like because it's too obvious that that that that that those are not real people.

They prefer the animal stuff.

if but what if we accept that those ads already exist?

I guess the fear is at what point do we stop recognising that the poor quality ads with fake AI people that, you know, kind of um are a bit jarring for us?

What what what at what point do we stop thinking that that's rubbish and do we just sort of accept that that

It is rubbish, but it's just there and it's everywhere and it's all around us, so you just try to ignore it.

And the sort of analogy I'd draw is I won't use reality TV because that's very controversial, but maybe something like maybe something like I'm gonna go out on a limb

here and think about things like plastic surgery and injectables, right?

Probably 20 years ago, people walking around with lots of injectables in their face, no matter how natural looking they are, was really shocking.

It's not anymore.

It's only shocking when it's really bad work.

It's only shocking when, you know, because it's just become an accepted norm.

And I think that's what the fear is, right?

That how long before we stop recognizing the poor quality stuff as poor quality, we may be just, it's not that we don't, it's not that we suddenly think it's good quality, but we

just sort of accept that it exists.

Because if we do that, perhaps it just takes over the majority of ad advertising inventory.

And before you know it, we've got rubbish quality everywhere.

But you're right.

Actually, it's a good point because the the fillers and Botox argument, there's a there's a massive online conversation at the moment around it started.

kind of it was it was hyped up with the Devil Wears Prada 2 coming out.

Yeah.

And doing side by side comparisons of the same actors versus them 20 years ago, where they look completely unchanged and they have not aged at all in those those pre 20 years.

And everyone was saying has has the emotion been been filled out of acting?

because it used to be that Philip, I think it's Emilia Clarke who says, don't use Botox before any acting gigs because I need the emotion in my face.

But there's so much work being done to people's faces now that you can't tell if there's no emotion being brought there.

think Lindsay Lohan was using an example because she looks absolutely incredible versus what she looked like 20 years ago.

But I think that that is a really good point because actually maybe that makes AI's argument easier because if there's no emotion left in anyone's faces anymore, it's easier

for AI to replicate it.

It is so true.

That's about the most dystopian

I do like to bring the dystopian element to the conversation.

You really do.

ah The second part to Barney's question was who's responsible for catching this?

I think we've answered that quite comprehensively that it's us.

uh It is.

And and and and that's what I mean when I say I'm kind of hoping that the test of this will be in the results of advertising campaigns, right?

And if they don't work, because if you think about what ads you actually respond to and what video ads, well let's even say TV ads that you respond to, it's stuff that's got

really good quality creative in it.

And if it doesn't have really good quality creative, you don't notice it, you don't remember it, you certainly don't interact with it.

I know there's a whole other layer.

you know, a subversive layer of advertising that's about brand recognition and all of those things.

But ultimately, we remember the ads in our lives that were of exceptional creative quality.

That's still got to be the standard, surely, because that's the stuff that moves the bottom line for brands and marketers.

But I think that also gets us into another conundrum, which is

What is it exactly that we're measuring?

And that's probably, you know, the the the sort of the crux of where we're at in the advertising industry, which is, and we talk about it almost every day, our measurement

standards that we invented at the beginning of digital aren't really keeping pace with actually what we need to demand from investment into advertising these days, because

actually the stakes are so much higher.

It's either privacy concerns or it's the quality of the creative.

So, you know, we need to be measuring different things and those things that we should be measuring is are these ads shifting more units?

Are brands and marketers selling more stuff?

Not just I'm a I'm a publisher or a content provider and I've sold the inventory.

It is has this actually made a difference for brands and marketers and their businesses?

And I would argue it's only good quality advertising that really does that.

I'm sure lots of people would disagree, but you know, that's well

I don't know if they, yeah, who would disagree though, because ultimately as you say, it's that, you know, money, it's down to what?

I don't know I'm trying to say.

I'm gonna say money talks, that's not what I'm trying to say at all.

But it's down to the bottom line.

And as you just said it, like if the money isn't coming in, if the revenue isn't coming in, then they have to revisit kind of what is causing that.

And well, let's have a look at our completely AI generated ads and see if that's causing the problem.

And every single bit of the campaign will get scrutinized and the creative is,

the first thing everyone goes to.

one of those things.

But but I also wonder if, you know, a bit similar to my example about, you know, Botox and fillers, I also wonder if the advertising industry still thinks that there is a high

enough tolerance in the general public for crappy ads.

Because let's think about the early days of desktop advertising and those ridiculous shoe adverts that just appeared across the top of your screen and chased you around for months

even after you'd purchase.

So, you know, i there'll be plenty of people that

have never even been exposed to that kind of advertising, but I think there is this thought, but particularly from a performance angle, that will tolerate quite a lot of

pretty average quality advertising.

And I guess that's where other technology tools and platforms come in, which is not only do you not have to you you have to have shitty advertising, you can actually use tools to

create better quality advertising, but in the long run it'll be better for your brand.

And who's your target market as well, because a lot of these studies look at Gen Z and Gen Z's reaction to AI ads.

You can imagine your older person, so my dad, for example, in his seventies, he might look at AI and he wouldn't have a fucking clue that it's AI and he wouldn't think it's

completely legitimate.

And so he wouldn't have any emotion over it because to him it's normal.

And so again, like where and how you use AI will come into play depending on what your target audience is.

I also want to verify the completely made up numbers I gave earlier.

I've looked them up.

The Kantar is Kantar's Media Reactions 2024.

And it's the other way around to what I said.

So 41% of consumers say AI generated ads bother them.

Whereas 29% of marketers say the same thing.

So it's not like AI ads, it's actually get disturbed by AI ads and 41% of actual consumers don't like them.

Whereas fewer marketers don't like them.

And so they need to kind of come up to the same level of consumers.

But again, it's still not a majority.

A minority of, well might be a majority depending on what the questions were.

But if it's a simple yes or no, then.

the majority are fine by them.

as DAIVID's own research points out, it's not gonna make or break a campaign, but it's how you use it.

And your anecdote, my mate, Jezza, uh is perfect for that.

It's exactly like we need the human touch to it.

And also we need to recognise that we have to be the ones putting that work in because there's nobody else gonna be doing it.

It's down to us.

It's down to like, if AI is going to really kind of work and help our business from a creative perspective,

then and from a performance perspective, we need to actually look at, need to scrutinize it heavily and see if it's working for us or not.

Yeah.

The ad industry seems to have decided in terms of demographics that my guy is Peter Crouch.

He is following me around everywhere.

Every advert.

No, just Peter Crouch.

you know, no disrespect to Crouchy.

Great footballer.

Great. Sure.

He's a great guy, but the company seemed to have decided 41 year old man must love Crouchy.

He's got a travel podcast with Abbey Clancy.

He's on the betting adverts on my sports channels.

So yeah, maybe they'll AI Crouchy at some point.

That's just made me laugh recently.

can, I really can.

Right, thank you very much Barney for that question, friend of ExchangeWire forever.

I'm going to move us on to question two from Kiessé Lamour speaking of friends of Exchangewire.

Indeed, absolutely.

We've had her on many times and she's wonderful.

So I'll get to her question.

If retail media is truly the third big wave of digital advertising, why are we still building it on first wave infrastructure incentives and measurement?

Ooh, Lindsay.

How'd you feel?

It's quite a sad question, isn't it?

Because I mean, she's not wrong, really.

The point is, um everything's being built on like first wave infrastructure.

And that is largely the problem.

I suppose retail media is getting scrutinized more because it is a newer channel.

And you know, as a newer channel, you should, if you're a new channel, everything should be able to be done with it.

You should be able to track every single thing.

This is brand new.

I'm investing my money into it.

I should be able to see every single little thing that's going on.

understand attribution completely and perfectly?

Well, no, because the technology that is built upon doesn't exist yet.

And that is the case with any new channel or any technology that comes in on legacy channels even, if you look at kind of out of home and if you look at audio, it's a similar

kind of problem, I think.

We're building on legacy infrastructure, we're using legacy measurement tools, legacy attribution tools, and retail media is a victim of that.

And one of the biggest problems I think that retail media is coming

coming under is it's probably the, to go back to obviously another conversation about AI, it's probably the main channel that exists is going to be probably kind of sort of in some

form or other, be completely subsumed by agentic.

The reason being that when it comes to retail media, if most of our transactions kind of end up happening through agentic platforms, and we don't actually interact with it, we

don't do any of the transaction ourselves, it's all done via agents.

You we had Kiessé's

colleague, Gemma Spence, a few weeks ago, a while ago, doing a coffee with me.

And we talked about this quite in depth, like, what does agentic retail media look like?

Who am I, if I'm an advertiser, who am I advertising to?

Am I advertising to the bot or to the agentic platform?

Or am I advertising to a human?

And that's a big, something we've got to wrestle with and reconcile, understanding kind of what that looks like.

I think that the problem retail media is going to have is, because it's going through that journey of like, some of it's going to be agentic targeting, some of it's going to be

human,

consumer targeting, it's going to really have a bit of an identity crisis, I feel like, and the technology it's built on is not going to help that matter.

you've got people in, know, retail media is a huge investment channel.

There's been so much growth in it.

And it's fascinating to look at how much growth we've seen across the world in the different formats that exist in retail media.

And you've obviously got not just with endemic brands, but non-endemic brands getting into it as well.

I find that fascinating.

It's such an interesting space to watch.

But it is, it does feel like it's sort of tailing off a bit because like the innovation can't happen, A, because of the infrastructure and B, innovation is not being stalled as

such, but it's being, but the conversation is being stalled because there's this kind of unknown when it comes to the role that agentic will play in retail media going forward and

commerce media going forward.

So I feel that that's going to cause.

a lot of questions to happen.

And VML, I know are doing some amazing stuff in this space and really interesting stuff and working with some really interesting clients, with big clients on things.

they will, they are almost, could argue, sort of leaders in this space when it comes to kind of from the buying side of things.

But if, you know, if even they're asking the question, it tells you that there's a lot to be, a lot of work to be done.

And the infrastructure isn't necessarily gonna change until, it's a bit like, we talked about this a few weeks ago with Omnicom's bet on agentic versus

ah you know what the likes of Index have done with containerization and and Bedrock containerization and it's like what do you need to get the the base infrastructure ready

before you try and build a genting on top of everything and retail media suffers here because everyone's trying to build a genting on top of it but the infrastructure is not

good enough to maybe do that and I worry about what that's going to look like in the next couple of years if the infrastructure remains kind of subpar and everyone's trying to kind

of look at what the next wave of retail media is because because it's the newest channel it feels like it should evolve the quickest

and it can't because the infrastructure isn't really there.

And I do wonder kind of what that's going to look like.

Yeah, I mean I think that it is definitely an issue of retail media building being built on top of existing ad tech infrastructure or legacy ad tech infrastructure.

So it's sort of mirrored um how we built ad tech for display advertising and search.

And I think some people would also argue, I know Kieron has said this, you know, certainly rowing back sort of 18 months, two years when retail media was really in a very explosive

s sort of state.

That actually isn't retail media just sort of a slightly fancier version of search advertising in an environment where there's maybe a little bit more intent.

Um, and he wasn't wrong because at the time, especially if you looked at retail media uh network owners, they were often retailers who were doing exactly that.

They're like, wait, we've got all this shopper data and we've got all this intent, so let's just find a way to monetize it and match those things together.

And their main aim is to sell their inventory, right?

On top of on top of being a retail media, you know, on top of

But potentially being a retailer, it's also to run their retail media network, right?

Actually the main aim for brands and marketers is to shift the bottom line.

And so that's I think where we're starting to see the disparity.

I think to start with, for lots of brands and marketers, it was worthwhile putting more investment into retail media because other channels were creaking, namely display.

If we think about all of the cookie shenanigans that went on for a good, you know, that that sort of a bit like COVID kind of erased a couple of years of our lives in ad tech.

Didn't it?

All for nothing.

Um, so you know, um during that time it was it was quite attractive to most, you know, intelligent marketers to think about shifting their budgets somewhere else where they

could actually see some results.

Retail media was the perfect channel at the time, right?

They understood the technology tools and and you know, broadly how the transaction was going to unfold, but it was a more intent-laden environment.

That worked to a point, but there is finite inventory available within owned and operated retail media networks.

And so then what do you do after that?

You need to expand your horizons.

And that's where we start to see the broader connection between retail media and commerce media.

And that's, I think, where you're starting to see, I imagined, a agencies working with their brand clients are probably asking questions like, well, how can I get this kind of

How can I replicate these sorts of transactions, but outside this particular retail media network, you know, how can I do this more broadly?

And I think that's a commerce media play.

Yeah.

And I think that requires more sophisticated, updated infrastructure.

And that's where the rub is.

So, you know, Kiessé's question is a good one because she's spot on.

We're still running mostly retail media networks using first wave infrastructure.

We haven't reached third wave infrastructure yet, but we're starting to see those new technology tools and innovations creep into retail media networks and other parts of the

advertising market, you know, other channels, to try to either replicate what you can get in a retail media environment, which is more intent and better data, or to try to improve

that retail media network environment and expand it out to a commerce media play.

But it's a really fragmented landscape, right?

There's the retail retailers that own retail, own and operated retail media network.

They're not going to welcome lots of new technology, depending on how what their tech setup's like, you know, whether they've white labeled it or whether they've booked

purchased something or partnered with somebody.

They're not just gonna open their doors to every other tech tool that comes along and they wanna be very um, you know, um protective of their data.

So

That's why you're starting to see, you know, we looked at um Kevel working with Nexta.

Nexta? uh thank you.

Yes, who they bought.

Yes.

It's just that all the ad tech companies got the same names.

And so after a while, you're like, Was it Nexta or was it Nexd?

Or was it, you know, Nex- something else?

Yeah, or Nexxen, that's it.

So, you know, Kevel's acquisition of Nexta was exactly that, trying to expand out that retail media concept and that retail media environment to a wider commerce media.

Yeah um sort of opportunity.

So I completely agree with you, Lindsay.

It is an infrastructure issue and an infrastructure problem.

And my take on Kiessé's question is, you know, the why.

Why are we still operating in that environment?

Well, there is a bit of a contradiction there in terms of you know, desired outcomes between retail media network owners and between what brands and agencies want.

That's number one.

They don't necessarily want the same thing.

They're not looking for the same thing.

And number two is it's a complex techno technology landscape and it's not as easy as going, okay, this this infrastructure doesn't suit our wider um, you know, desires.

So let's just add a few extra things on, or let's go into another network environment and see what we can get.

It's a case of kind of

mixing things up and seeing what works best per client, which is probably where Kiessé's frustration comes in because she's working for an agency with very progressive brands who

want more sophistication.

And I think what I'd say is that's coming, but we need to start thinking about things a bit differently.

And for you know, because brands want an omnichannel approach.

So they don't then they don't want to just be locked into walled garden after walled garden after walled garden.

So that's where the retail media industry, I think

has got a bit of a reckoning coming its way.

You know, th they need to think about how they can either open up their networks to enable brands to bring in more data or other technology solutions to help activate that data, or

they need to start thinking with a commerce medium.

Yeah, I agree.

think something you sent John was a quote in it from somebody in the US, tech exec in the US who said that in the world of agentic and in retail media, only Amazon wins.

I think that that's a shame and maybe a bit reductive to look at things that way.

we do probably, and everything you said, Rachel, I think highlights, we do probably have a bit of an ownership problem.

And as we tend to do in the ad tech industry, there is an ownership problem because like...

No one owns retail media.

Gone are the days where you'd have BOK and AppNexus kind of building something and then they take ownership of the ad server, they take ownership of the SSP and then everything

is kind of built from that one thing.

Things aren't like, there aren't as many standards in retail media.

I think IAB, what have I got here?

IAB Europe have released the Commerce Media Measurement Standards V2 with Tesco Media, which will set a default 30-day look back window and it's underpinned by a retail media

certification program.

which has retailers demonstrate audited compliance with a transition window to end of July, 2026.

That is something and that's going to make a difference when it comes to the measurement and attribution side of things.

And it's a starting point, it's still, know, compared to kind of how big retail media is and the direction it's going, there's going to be a lot more required and people are going

to need to feel they can turn to someone.

And I think, you know, obviously, IAB's tend to sort of set their stalls out with these things.

I think they're going to be the people that are going to have to help the industry out here when it comes to standardization and when it comes to regulation and understanding

measurement.

Because as you say, Rachel, like it's all outcomes led, it's all bottom line led.

If I can't measure that, then I'm not going to invest in it.

I'm just not going to.

Exactly.

At some point, that is what's going to happen, right?

And because I didn't even touch on the AI and the agentics side of what you were talking about, Lindsay, that every time I go back to that mentally, I just think about how people

discover products and information in an AI search world.

What it means is that brands, all brands and publishers and content creators need to

have a brand and understand how they show up and are visible in an AI search environment.

And that completely changes the landscape.

And so Lindsay, you very articulately gave us a sketch of what that looks like from a tech perspective, which is effectively at the moment still a bit of a mess because people are

trying to build on top of existing infrastructure.

But it's even more urgent for retail

media network owners, especially if they're owned and operated retailers and any brand out there and also any publisher out there, because you need to know and understand how you're

visible in that environment before you can even be offer thinking about how you offer brands and marketers some kind of value for advertising within any kind of environment

that you've set up.

So it's far more complex now when you think about um AI replacing our search habits.

So

It does feel like the agentic AI onslaught really did pull the rug under what we thought retail media was going to become.

Yeah.

Totally.

And and and and I go back to, you know, what I said at the start of this, which is Ciaran used to talk about it all the time in the early days of retail media.

it's just slightly more sophisticated search advertising.

Well, you know, we're in a new world.

Search doesn't exist anymore, so we're gonna have to think about it again from a completely different perspective.

And that's why, again, retail media network owners need to get more commerce media focused in their outlook I think.

Great, excellent.

Right, thank you very much, Kiessé, for that question.

I'll move us on to our final question of the mailbag from Kamel El Hadef from Audion.

So his question is, AI agents are enabling audio advertising to evolve rapidly.

Brands increasingly use it to drive tangible outcomes as well as build brand awareness.

But are marketers ready to adopt audio as a full funnel channel?

I love audio.

um Lindsay, what do you reckon?

I also love audio.

Audio, think, is an underrepresented channel.

Agreed.

um And it's not talked about enough and it feels like it's often talked about in the audio echo chamber and there's not enough conversation outside of that echo chamber about how

great audio is and can be.

One of the biggest problems with audio is and still is measurement.

And I think that that is where it falls down as a full funnel channel.

And, you know,

We listen to us, we're recording a podcast.

We obviously, we have an understanding of audio measurement ourselves in terms of what we can see, what we can't see.

Audio measurement is limited and it requires a lot of understanding.

There's very little last click attribution abilities with audio.

You're still having to of understand, understand it with the context of other channels and use it with mixed marketing, immediate mixed, I can't even speak, MMM.

and attribution models, like that's where it really kind of shows its value.

As a standalone channel, it's much more challenging to be able to see that.

And there's definitely improvements, definitely things are being done.

you obviously, Kamel's question around kind of is, is agentic, is audio, my God, I need to go have a lie down.

What role is AI playing in improving this?

Like it's obviously playing an absolutely huge role and it will continue to, but it's still,

I don't think is, it's not necessarily ready in terms of from a marketer perspective, well, a marketer wants to break out audio and go, do you know what?

I feel like I know what I'm doing here.

I'm to put everything in audio.

think from the, as a performance channel, I'm not ready to do that yet because you just can't see it.

I think as a brand awareness channel, think brands are all in, but when you're trying to look at it, you can argue that everything is now performance branding or branding

performance.

And that is very, very true.

And obviously from an outcome perspective, audio plays really well along to other channels, but

as a standalone performance, lastly attribution channel, I don't think it's there yet.

Or certainly marketers don't see it to be there yet.

And marketers are resistant for that because they need to see that information.

Tied with everything else we talked about this episode, the measurement piece is what's often lacking and it's so important.

m there's a walled garden issue here within the audio space that it is difficult to get data out of those walled gardens.

So you are having to use different tools to try and pull that together and...

different platforms report on different things.

They'll report on listens, they'll report on engagement.

And so you can't even pull that together to be able to understand kind of what that, what, you know, my ad on this radio channel, on this podcast actually drove.

I think it's a challenged channel, lots of opportunity for it.

But I also feel like it's catching up, but I feel like audio, I think the audio industry want audio to be seen as a performance channel because everyone talks about performance.

But I think also it should kind of like,

almost accept and embrace that it's a really strong brand awareness channel and lean into that more because it really is like, it's one of the most uninterrupted, you know, I.

It's an entertainment channel.

Yeah.

One thing I can't skip, like I'm normally walking or driving while I'm listening to all these endless ads on the podcast listen to, like I can't skip them and with radio I can't

skip them at all.

Like with everything else, I will skip it where possible or I'll walk away or I'll turn away or tune out.

In my ear, I'm not tuning out.

So I think they need to really lean into the power of that in absence of better measurement.

Yeah.

I mean I feel like we have we should have this conversation with a marketer to really understand what it is that puts them off seeing it as a full funnel uh marketing

opportunity.

Because I think theoretically you're absolutely right, Lindsay, but I just wonder how much of that is um a a perception, because I don't believe that of all the, you know, of all

the sort of audio tech solutions out there.

they're not putting effort into making the measurement part of it a little more uh you know a little more detailed, a little more specific and improving it.

They have to be, right?

Because the other thing is as well, I no longer see audio as just an audio channel.

It's also a video channel because what's one of the main drivers of audio?

It's podcasts.

What are we doing right now?

Filming our podcast.

A mate of mine who works in post production, Cheryl said, it's basically the chat show.

It's the chat show reinvented.

The chat show's having a massive renaissance.

That's it.

So it's not just an audio channel anymore, but if you are a podcast, regular podcast listener, then you depending on where you consume your podcast, you're getting it via

audio and video.

But that then, if you're a marketer, then, and any media buyers will agree.

that then becomes a challenge of, well, what am I buying?

I'm buying audio.

audience don't want to hear about it being, the podcast being viewed on video because that's warped, that's So, like, it's completely separate and that's it becomes a massive

challenge. And then it's just like, who buys it?

And how do we then bring across the two different departments?

So we now have the AV team and the audio team, like,

okay, the podcast performed this way, the video version of the podcast performed this way, we've got to bring those things together and attribute that correctly.

Like it's made everything a lot more complex from a marketing perspective.

But that's part of the problem, right?

And that's what I was going to get onto is that, you know, budgets and the way that it's bought are still very siloed.

And I just think we are really moving towards a media environment where we can't afford to be so siloed anymore because everything is potentially cross-channel.

And that in fact is not just the way that we're consuming things, but it's also the way that AI and generative AI is driving us.

And also ultimately, it's obviously better for brands and

I'm going to scoop up publishers in that too.

If they're being discovered across a whole range of different channels, that strengthens their brand.

I understand there's some complexities from a tech perspective and also from a measurement and attribution perspective.

That's exactly what we've got to get on top of, right?

And I know that's not, you know, as easy as just making those statements.

Yeah.

But I feel like, you know, and we've been talking about the siloed approach to uh, you know, marketing budgets and what happens within agencies and various buying teams.

For as long as I've been running ExchangeWire.

So it would have been going on for another 30 years prior to that longer.

But, you know, I think there is now a real impetus for everybody to change because there's only a finite sort of amount of advertising and marketing dollars out there.

We're not looking, you know, we're not looking at an industry.

There's still a lot of growth, obviously, in the advertising industry, in the digital advertising industry, if it even needs to still be called that anymore.

But we're not looking at the same growth trajectory that we were looking at at the early days of programmatic, where it was considered a specific separate channel.

It's not like that anymore.

Actually, what marketers want to see is a change to their bottom line, and that's where investment is going to go.

And therefore, I know that most agencies are wanting to drive their clients to look at that across the piece from a cross-channel perspective.

So it's got to be possible to do it.

I think I I be UK tracked ad spend for audio as growth grow a 10% growth in 2025.

So there's growth there is massive.

There's definite growth there and there's definitely opportunity for it.

I think that a really good example touching what you were saying, Rachel, is what we see from media owners and tech suppliers in the out of home space.

The buy, not the buying of out of home is siloed, but the buying of different channels are still quite siloed within agencies.

not, and that's a blanket statement, so don't come at me.

But em there is still lots of silos when it comes to certainly these channels like audio, out of home and AV more generally.

But what we see from media owners and tech suppliers in the out of home spaces, they do not want to have an out of home conversation in isolation.

Exactly.

It's having chats with them.

They want to talk about out of home in the context of omnichannel.

And I think audio need to kind of do the same thing and kind of go, yes, I'll talk about audio uh on this podcast, on stage, whatever.

But I want to have the conversation as it pertains to omnichannel because that's where it should sit and should exist.

And all the out of home media owners, this applies to London, it applies to Singapore, it applies to MENA as well.

Any conversation I've had certainly.

personally about out of home has been like how it fits into everything else.

I don't want people just kind of see this as another out of home conversation.

I want them to see how it plays into their omnichannel role because that's what they're trying to kind of get agencies and buyers to see and audio needs to exist in that same

space.

Yeah, agreed.

Sort it out, audio industry.

I think that was a rich and fulfilling discussion around audio.

I'm going to go and listen to a podcast and buy whatever Peter Crouch tells me to buy.

uh

interested to know what he advertises.

He does a travel podcast.

They can answer your travel queries on their lovely couples podcast, Abbey Clancy and Peter Crouch.

No, but I'm advertised it every day.

Well, that was a lovely episode.

Thank you both so much for joining.

Thank you.

Lovely to be part of it.

And thank you to Kamel, Barney and Kiessé for your questions.

I'm going to pop their LinkedIns in the show notes.

So if you want to pick up any of these issues with them, feel free, but thank you for contributing and thank you all for listening and watching.

And watching.

You've been listening to the MadTech Podcast, brought to you by ExchangeWire.