Your Exclusive Shoptalk Europe 2026 Key Takeaways Podcast

Okay.

Hello, everyone, and welcome to your shop talks Europe 2026 key takeaway session.

I'm Chris Walton.

I'm the CEO and President of the Omnitalk Retail Podcast Network.

And you all in the audience, in addition to the three panelists next to me, are about to be part of a live podcast, which means that everything you see and hear on stage over the next 40 minutes will be available via the Omnitalk retail Podcast network, which you can access at any time at your leisure following the conference, wherever you get your podcast.

So joining me now on the far end is Rebecca Bemhyna.

She is the Shoptalk VP of Content for Lux.

Welcome, Rebecca.

Thank you for the welcome.

Thanks for having me.

Lovely to be here.

Yeah.

All right, and then we have Adam Plarm.

He's the Shop Talk VP of Content Europe.

Hello.

Good to see you.

And then we have Ben Miller, of course, to my right, shoptalk's VP of Original content.

Good evening.

All right, let's get to it.

All right, so here's how the next 30 minutes are going to work.

I've selected seven topics.

I'm going to allot no more than four minutes, give or take, to each topic, and that's gonna keep the conversation moving, or at least that's our goal here.

So let's get started with topic number one.

Ben, I'm going to you first.

AI.

Where are we at with it?

Coming out of the conference, what's your thoughts?

Okay, I mean, I think it's where we had to start.

We're on sunset stage.

This has been our AI stage for the show, and it's been packed.

I'm sure many of you have been here.

It's been standing room only.

At times.

The energy has been about understanding where we are.

So I think what I want to talk about, because I think there's a lot we could cover, is agented commerce.

So where are we right now in this moment?

In Europe in agentic commerce?

I think it's really clear there's two things going on.

One is agented commerce, as you might define it, as pure agent to agent automated.

We're not at day one yet.

Nothing.

Zero hasn't started.

It's still hypothetical.

So therefore, all the conversation has been about agented commerce, as in, how do you use answer engines, Gemini, ChatGPT, whatever you call it, how do you use them for commerce?

What is the commerce opportunity, and how do you get your products found?

What's super clear is in the US because of protocol rollouts, whether that what Google's doing with Universal cart, what Amazon's doing.

So you look at the new Amazon Shopping Rufus product as was, the rollout is six to 12 months ahead of what we've got in Europe.

So tangibly right now in Europe, some of these agented commerce tools aren't available.

So I think a conversation has been for people in global roles.

It's okay, what are we doing?

How are we getting ourselves up?

For people with European businesses, the energy of the conversation has been around how do we find, how do we get our products discovered?

So all things around Geo SEO, that's been kind of the hottest topic, but in the knowledge that this next level.

So use the agents to buy.

It's coming because we can see what's happening in the US where some of these rollouts are already happening.

But in Europe right now, it's all about discovery.

And the numbers show that shoppers are using it.

Some of the numbers that have been shared have been huge, but we'll come on to that.

So really the focus being step one.

Let's focus on step one, particularly over here in Europe in terms of getting ready for discovery.

Yeah, I think we're probably at step two.

Step one is recognize the theory.

Step two is start doing things, start doing stuff on discovery.

Step step three will be buy, and then step four is automation.

And I think in the US we're starting to look at three, and I think we're firmly in two.

Okay, well articulated.

Adam, what do you think?

What do you add?

I think just to take maybe a gentic out of it, I think for me, more than any show we've had in the last couple of years, the focus is just purely on execution.

There's so much talk around, okay, where are we getting ROI from our AI projects internally?

And the consensus seems to be it's operational still.

That's where people are getting the most value.

I think we heard from Jose from etam.

He said something like, reality is efficiency for AI right now and autonomy is hype.

And I think so much of the value that's coming through is in operations, in supply chain, not so much customer interface still.

Yeah.

So, Ben, let me press you too.

I want to go back because Adam brought it up.

The operational side of it.

Yeah.

Does that framework still hold on the operational side?

Because it felt like that was more of the consumer focus that you were talking about.

You're absolutely right.

I mean, that for me was where we are with agent E Commerce, customer facing.

Yeah.

I mean, there's been so many use cases.

We're talking AI broadly.

So Many use cases around operations, around efficiency.

So just to give you one example, a conversation, a panel that I had with Amir rashkar who runs negative360, that's Sainsbury's retail Media network.

They have rolled out a tool which allows a level of self management of, of your ads within the network using generative AI.

Once you are happy and a creative has been accepted, they use generative AI to tailor the image to all the different places where you would shape sizes, formats to run the ads across their whole network.

They've taken a process that would take three to four weeks, is now done in 90 seconds.

So those case studies are live and happening now.

Yeah, that's why my prediction is actually that next year, this time at this conference, we are going to have a definitive session on how do I actually manage an agentic employee.

I think that's my prediction for what Adam, you're going to program next year.

Okay.

All right, well, let's keep moving.

Adam.

The idea of this was the retail festival where AI and human ingenuity take center stage.

So my next question is human ingenuity, where are we on that?

How's that coming out?

Yeah, it's really interesting.

Referenced this in the opening remarks.

I think on, on the first day, it was something that our audience, our community was so passionate about.

It was yes, we want to hear about AI, of course, but we need to be balancing that.

And I think there was two, two strains to that.

One, it was, we want to hear about how AI is empowering humans, whether that's colleagues or consumers.

But more than that, it was, yes, do your AI stage where we are now.

But, but amplify the content on the role of stores in the age of AI, if you like.

What role are stores playing now for loyalty, for community, for connecting to cultural moments, temporary retail spaces versus flagships.

What's the conversation there?

Brand storytelling, sort of emotional brand marketing side of things.

This is all the stuff that our audience, I think, and our community are very keen, was not lost as a result of the deluge of AI content.

And so that came through, at least for me.

And my conversations in the sessions I saw was a really, really strong, A really strong theme and a really strong counterweight, I would say, to the AI conversations.

So.

So yeah, we heard a lot of time, we heard a number of people say that AI is going to make us more human.

Like makes, makes our interactions more human.

Rebecca, what do you think of that statement?

Yeah, it's an interesting one.

So I heard from Amish, from Boots, who was Speaking on a session in here actually around visibility in the age of AI.

And what he was talking about actually was the most high value use case that they're seeing in Boots in terms of the use of AI within their workforce is equipping their sales associates and their store associates with AI to basically free them up to be able to have that more human interaction with their customers.

And we hear about this a lot in luxury, particularly if you think about clientele.

And it was really interesting though to hear from a kind of non luxury retailer how they're thinking about it.

And it sounds quite similar actually.

So really focusing on kind of driving efficiencies, using AI to kind of automate some kind of boring, I guess, manual tasks that now can be done with AI and really allowing for that, that human interaction and human kind of white glove service, if you will, which they talk about in luxury.

So I thought that was really interesting.

Yeah.

So you buy that sound bite then that AI is going to make us more human in our interactions.

You think that's a, a true statement versus like a marketing pitch to make it sound like AI is something we should deploy in our organizations.

I think there's definitely some, some use cases where it will.

Yeah.

Where we sales associates, store associates will have more time ultimately to be able to spend with the consumer.

And it's.

And that's thanks to AI driving efficiency.

So I think that.

Yeah, I think we can say that.

Well, I'm going to defend Rebecca's stance on this one as well.

There was a quote from one of our stores sessions earlier in the week that was huge.

Human first and digitally connected or digitally enabled, something to that tune.

And again, it's the same thing.

It's freeing up mental capacity and actual time for associates to be better at the bit that can't be done by AI and tech.

Yeah.

As much as I'm pushing you both, I actually agree too.

I do too.

I mean, I think history is our guide.

Right.

Like psychologically, like even if you look at the advent of social media and how that came about, you know, all the algorithms that went into putting content in front of us, we want that personalized because we want the better deals, we want the better products.

And so this just amplifies that too, in my opinion.

All right, number three, most impactful topic or trend that actually isn't AI.

Adam, let's go back to you.

Yeah, I dominated.

One thing I found really interesting actually was it's a bit of a cliche quote at this point, but everyone talks about the consumer doesn't think about channels.

And that's kind of a statement that people say.

But actually I felt it was coming through quite practically in a lot of conversations throughout the show around brands and retailers showing up in places where maybe they wouldn't have done historically because of perception or even their own view of that's not right for us.

We shouldn't be doing that.

And a good example was Nadine Graf, who's a regional president at Estee Lauder.

She said, you know, we're on Amazon because our clients are there, our customers are there.

The only thing that held us back was a perception of premium beauty that we had, but the customer doesn't have that.

So if your customers are on Amazon and you want them to buy from you, you need to be on Amazon, which I found interesting.

And we kind of joked about it being like channel pragmatism.

But there was a few others as well.

You know, John Lewis, five years ago, I don't think you'd have associated them with being on TikTok shop.

You wouldn't have associated them with.

With being on things like Just Eat and, you know, that kind of thing.

And even Zios, sorry, Zalanto and Vestia last week, a few years ago, would that have sounded right?

And now it's like, yeah, okay, I get that.

That makes sense.

So that was really interesting.

And that was coming from very senior level conversations as well.

Yeah, yeah, my mouth dropped when she said that.

Like, I was just shocked at how unashamed she was about talking about how successful they're being on Amazon at Estee Lauder.

I think.

Can I jump in as well on this same element of channel pragmatism?

I did a conversation yesterday with Carlos, who is the CEO at El Campo, and we talked about the changing Spanish grocery consumer.

So El Campo Oceans business here in Spain, it's a hypermarket retailer at heart.

And we talked about all the demographic changes, smaller households, longevity, declining birth rates, people shopping more often, buying smaller portions.

All of which is headwind stuff for a hypermarket operator.

So in response, they bought a convenience store business, rebranded it, integrated it the partnership of Ocado to up their game in E Comm.

My challenge to him was, great, you do all this to follow the shopper, but really all you're doing is adding complexity into your business.

And the number one enemy of a grocer is complexity.

Yeah, when you've got 2% margins to play with.

And that's a general, not an El Campo number.

And his pushback was really interesting.

It's not complexity at all.

Grocery is a really simple business.

You've just got to buy good stuff and you've just got to get it to your consumer.

And that's really interesting because we've been talking so long about individual channels, their profitability, how you think about it.

So again, this channel pragmatism was not something I was expecting to hear much about at this show and it keeps coming out in different ways in different places.

Yeah, that's interesting too when I start thinking about how grocery appears to be leading that discussion even globally now in terms of how retail is going to play out.

As retailers grapple with this question of where do I show up, how do I show up, how do I work within the context of AI.

Rebecca.

What?

Anything you'd add here?

Yeah, definitely.

I'd like to shift gear a little bit and talk about something quite different.

This is a bit more strategic, I guess than what you've both been speaking about.

Well, no, as in this is quite tactical.

This is a bit more kind of strategic, higher level, I would say.

Love it.

So there was a globalization session that was on the agenda earlier today, which I'll talk about in a moment.

But then it was interesting that the theme of globalization came out in some of our keynotes a bit more organically, which I thought was interesting as well.

So it's obviously a big theme high on leaders minds.

And in the globalization session they said that three quarters of international brand expansions are now failing.

The western brand halo effect has diminished.

Local challenger brands are now competing due to really deep cultural understanding.

And we're seeing in China, we're seeing brands like Songmon Icicle as well, which Kering has just bought a minority stake in doing really well and you know, locally made brands produced and really speaking and tapping into that local culture.

And so I think the kind of takeaway here is just underscoring the importance of thinking about local teams, local insights and local execution and really crafting the assortment, the stores as well.

Nadine was talking about the assortment.

We had Alexi Muro from Christine Louboutin who was talking about how all of their stores are completely different in each market.

And that's obviously very intentional.

Their merchandising is very different and assortment is very different as well to try and drive their customers to visit different stores.

And another just lastly on that globalization point, what Nadine said I thought was really interesting because you often think about western brands going into new markets, localizing, adapting their offerings for that market.

But what she was saying is they found in India.

So she's responsible for.

I think it was 90 markets.

Is that right?

And in India Kohl.

So the type of eyeliner is really popular and they are starting now to kind of bring that to markets outside of India.

So it's almost like the kind of full circle, if you will, of kind of insights, execution and kind of really thinking about.

Yeah.

What.

What different markets can bring to other markets in the world as well.

So I thought that was super interesting.

Yeah.

Localization always a top priority for.

For, for retailers.

All right, next one.

This is a fun one.

Stat or slide of the show.

Rebecca, let's go to you first.

For me, it has to be 90% of Zalando's online content is AI generated.

When David Schroeder, who spoiled talk on our main stage on day one, that blew.

That stat blew my mind.

And it got me thinking of a lot of questions that I would love to have the opportunity to ask him.

First of all, why 90%?

Why not 100?

Is that intentional?

Is there 10% that they are intentionally not making AI generated and why?

And I think the other question that that raised for me is if companies are able to scale up your content production to that level that quickly as well and another can follow sweeps kind of what then becomes your differentiator?

So is the human.

And back to the human ingenuity is kind of human kind of guts.

You know, if you're thinking about the taste and judgment, I suppose it comes down to that.

So I think that has like that really left me with a lot of questions.

That's that and I think it's super interesting.

Yeah, it was a great stat.

Like, and then the starting point too was zero.

Right.

Didn't you say it was zero to 90% in a really short period of time too?

I don't remember the exact time, but.

It was a year.

Right.

That's just, it's wild.

And it also starts asking the question too of like, what are we okay with that's AI generated as consumers?

Because it seems like he's not concerned about it.

Right.

You know, Ben.

Okay, I'm going to change of tact.

I'm going to share a stat from one of our research presentations.

On the opening day, we often do a series of scene setting presentations.

I think the person who gave the presentation is in the room.

So shout out to Rhian from IGD.

They shared the forecast.

We talk about GLP1s.

Some of you might have heard Adam and I do as zeitgeist earlier.

We have this belief that the impact that GLP1 medication will have on retail is going to be bigger than people believe right now.

It's going to be huge.

During the zeitgeist, we shared a stat that we've used before, which is some Morgan Stanley data, which is they currently believe that 1% of the world's population is on GLP1s and that's going to grow to 10% in a decade.

Rhianne's number was talking about the UK market, where they've done a really deep dive.

And their forecast is that by 2030, which really isn't far away now.

No.

Which is slightly frightening, 15% of the UK population could be on GLP1.

Wow.

15%.

Remember, if you're on GLP1, you question everything about your food consumption.

It's like a switch that happens.

Nearly half of people that go into GLP1 will change their main grocery shop.

And those people in the grocery.

You know how hard it is to get someone to change their main shop.

So I still think we've underestimated it.

And if we've got 15% in the UK, you know, we're going to expect to see big numbers right across Europe.

And that.

That was a stat that really stopped me in my tracks.

That's interesting because usually in this session I always ask, like, is AI underhyped?

Overhyped, Properly hyped?

Yeah.

So if I was to ask you about GLP1, you'd say it's still kind of under hyped?

Yeah, absolutely.

Okay, Adam.

I'll stay on 2030 vision.

This was shared by Holden Bale from Merkel on day one.

It's my favorite stat of the show, but I also don't necessarily agree.

Oh, I can't see it.

But.

But that, you know, who am I to do it?

He said between 10 to 25% of global E commerce, we fully automated by 2030.

I think there's a big gulf in 10 and 25% to start with, but also that it does sound a little.

Bit of a hedge.

Fair enough.

And it sounds high, but the pace of change and the way things are heading.

Maybe, but it stuck with me because I didn't agree.

Sorry, hold on.

Yeah, yeah, you guys all gave stats.

I'm going to give a slide.

Okay.

Okay.

I'm going to give a slide.

My favorite slide was from Tom Cooper, who's the CEO of Duvo, and if you don't know him, he was the former founder of the Roll It Group, which is a very successful E grocer.

And now he's got the AI bug and he's created an AI startup all around agentic AI and bringing that inside of organization.

So going back to what we talked about in the beginning and he had this great slide where if you can get your hands on it, I highly encourage it.

He said, basically, Here are the 10 areas where you could deploy Agentic AI today in your organizations and immediately find money.

And it's things, things that aren't sexy.

It's like following up on vendor compliance issues and things like that.

That was one of the ones that stands out in my mind.

So it was just a great slide that encapsulated everything you can do very quickly by thinking about how do you deploy agents to your organization.

I'd say vendor compliance is quite sexy, surely.

Yeah.

Oh yeah.

I mean, yeah.

No offense to the vendor compliance auditors out there, Adam, but.

Right.

Well said, well said.

All right, quote of the show.

Ben, let's go back to you.

Yeah.

So I spent a while in the various retail media sessions, a subject that continues to evolve and it's huge for many of our community.

And there was a great quote from Thibaut, who's chief operations officer at Unlimitel.

So Unlimitel is Cathor and publicist jv and we would talk about the relationship between a retail media network and the retailer that they're serving.

And it said the relationship is best when it is autonomous but not independent.

What I mean by that is you need the autonomy to create the culture of an agency with a separate P and L and the ability to invest and the ability to make decisions that are not just bound into a JBP somewhere else, but they're not independent.

The success of the media network depends on the success of the retailer that they're part of.

And the best results come when you can get three parties round the trade partner, so the advertiser, the retailer themselves, to make sure you've got the right assets available and the network together.

So autonomous but not independent.

I really like that.

Wow.

There's a lot to unpack there.

I might have to do a separate podcast just on what that concept means.

Ben.

Love that.

Wow.

Rebecca.

Yeah, I would say this isn't a quote verbatim, but I really liked what Alexei shared during our main stage keynote.

He I thought it was really transferable to any brand in any other category as well.

It's not just luxury specific insight and particularly for brands that are founder led and they're trying to make sure that the brand outlives the founder as well because Lots of decisions for founder led brands are typically gut based and what they have devised is this kind of brand pillar framework, I suppose.

And it doesn't just sit in a kind of internal presentation somewhere on their system.

They actually use their brand pillars as a filter to decide whether a new idea innovation, whether it's a client collaboration, a new campaign idea goes through to production.

And so basically they have their five brand pillars and if an idea doesn't hit at least three of them, doesn't fulfill at least three of them, it's a no.

And I think that's a really smart, really structured way of kind of building a kind of decision making framework to enable decisions and to enable decisions, I guess, faster as well.

Because gut, I think, well, you run the risk obviously of, you know, if the founder or, you know, leaves the business or you know, you want, you want to create a brand that outlives the founder.

And yeah, I just think it makes decisions faster as well to enable them to be faster if you have this really clear framework.

So I love that.

So you like the framework that they were espousing?

Yeah, yeah, I did and I.

Well, what their framework is very on brand and I mean, I think you understand the brand more when you understand what their brand pillars are.

I think you understand the decisions that they make, the campaigns that they run because you see it come alive, you see those brand pillars come alive.

So yeah, it made me understand the brand a bit more, I think when.

I said, yeah, it's a great point because it is important to have decision rules around what we're not going to do as much as it is important to have them for what we are going to do.

Right.

Adam, the one that I really like, I don't think it's a direct quote actually.

It was a key takeaway from the end of a session today where we had Pinterest, TikTok shop and Reddit talking about kind of new routes to discovery and the collision of Entertainment and Commerce, etc.

And the line was that confidence is the new currency.

And I really like that because it was, I think we talk a lot about trust and authenticity and all that kind of stuff and actually confidence feels like the next layer upon that.

And it's like if as a consumer you go onto these platforms and you see a new brand, you see a product from a brand, you know, or maybe not.

And if you feel really confident that it's, it's right for you because of the type of content you're seeing, because of the, you know, some of the UGC or whatever it might be, live shopping any of these channels.

I don't think it matters what the platform is.

Then you're far more likely to buy, obviously.

But obviously the loyalty is there as well.

I don't know.

I just really like that confidence is a.

As a currency.

I could tell you liked it because you kind of leaned back in your chair as you said it.

As I almost fell out of mind.

Trying to mimic you, I made me feel confidence.

Yeah.

Yeah.

Confidence is a new currency.

Yeah, I like that.

I like that.

Mine was.

I had the chance to interview All Saints CEO on stage on Tuesday, almost forgetting the day now, Tuesday.

And he said something to me that I've never heard before that has stuck with me and I actually find it really inspiring.

He said, the CFO's job is to ask what can go wrong, but the CEO's job is to ask what can go right.

And that's where he, as a CEO is focused on in terms of how he communicates with his board, how he talks to his team.

And I think if you take that approach in your day to day work and interactions, it's going to lead to a more inspiring environment.

And so I was pretty impressed by that.

Yeah.

All right, next one, best offstage insight.

Let's go to Rebecca.

Yeah, I had a really interesting conversation where agents were referred to as boring shoppers.

So agents are not browsing for inspiration.

They are looking for the fastest path to the product that kind of best matches a set of really specific product metadata.

And so therefore kind of having high quality product metadata may be a determinant or one of the biggest ones actually, of whether your product gets recommended at all.

I thought just kind of referring to them as boring, boring shoppers, I thought was quite.

Yeah, it was quite a funny image.

That's definitely stayed with me.

And I think just that that kind of outcome, or rather takeaway from that, is therefore brands need to be really focused on product data.

I think that was one of our takeaways from the European Retail Tech Investor session that I moderated on Monday.

We had a partner from Carl Square, which is an investment bank, and he.

He fundamentally believes that brands really need to be thinking very deeply about their metadata and making sure that that's all in order, in order to be able to be visible and be able to participate properly and drive commercial meaningfulness from agenda commerce.

Yeah, the boring thing gets me thinking about Maya too, because I had this great conversation with the founder of Zenline AI, where they're looking at how do you deploy Agentic agents into.

Into your.

Basically your merchandise planning organization and your merchandise assortment decisions.

And he said something similar to me.

It made me think of what you said.

He said an agent inside your organization is like your employee on day one.

People have the idea that we can just take the LLMs and all the information and data they can provide to us and just start using them.

That's not true.

The way you have to think about it is if you had an employee on day one, would they know your company?

Would they know its language?

Would they know its processes?

Would they know its data?

Absolutely not.

And so that's how you have to think about your agents as you're starting to deploy them and set them up for success.

Because then once you get them in place, not to say they're boring, like actually, the newest employees often are not the most boring.

But that's a good way to think about it mentally in terms of how we need to start managing agents going forward.

Ben, Adam, anything you want to add?

Ben?

Yeah, a quick build on that one.

And it's probably sometimes when we do these sessions, we talk about what haven't you heard as much of as you expect?

And that whole area of the context layer.

Maybe I've been in the wrong conversations, but I've not heard that much of the show.

And I think we're increasingly realizing that you need that context layer to be able to apply a lot of the tools.

So I think that's super interesting.

My favorite one was something that somebody said to me Tuesday evening.

We were talking about E commerce, and this is somebody who manages a very large E commerce business.

And he said, you're thinking about it the wrong way.

E commerce is about mastering algorithms.

So if you master an algorithm, the rest will follow.

P and L people structure the first thing.

The most important thing you should be doing is mastering algorithms.

And what we're seeing right now is a lot of new algorithms that we don't quite understand.

So that's creating definite headwinds.

But I think it changed the way I thought about it.

Yeah, it changed the way I'm thinking about it too.

Wow, that's a very heady one.

Again, Ben, thank you.

You got my head spinning on that one too.

All right, Adam, what do you think?

I'll keep it light.

I had a great conversation on Tuesday night with a head of innovation who remained nameless.

But if you were in the session before, you'll probably know immediately who I'm talking about.

He had some very, very choice words for innovation labs, innovation factories, and his summation was that if you operate in that model in this day and age, you're just performing innovation theater, which I really liked because I always thought the concept of a factory in a lab sounded quite good.

So it kind of challenged my perspective.

And you're nodding, I guess.

That's your.

Yeah.

And Ben, roadkill on the side of the road.

Yeah.

No, I think I probably agree with him.

Yeah, for sure.

But then going back to Ben, Ben's your point, too.

The interesting thing about what he's saying and what you're saying is generative AI is not algorithmic.

That's kind of the beauty of it, too.

So where does the rubber meet the road on that in the long run?

Yeah, and that's exactly what I'm going to talk about in the next question.

Oh, really?

Exactly.

Wow.

It's almost like I knew that, but I actually didn't.

I had no idea.

All right, what is the biggest epiphany then?

Well, then I'm going to hold that for.

Then I'm going to come back to you.

What is the biggest epiphany you had at the show?

What do you know now, leaving the show that you didn't know before going in.

Adam, why don't you go first?

This one's actually quite narrow, but it really stuck with me.

And I sort of did one of stats from Holden's presentation, Holden Bell's presentation.

I didn't agree with this.

I did find very thought provoking.

He shared a slide that suggested there's a correlation between angst in the workplace over AI tools and workplace rights within that country.

So the stronger the worker rights in that country, the lower the angst about AI in the workforce.

And obviously then the reverse.

And when you say it out loud, it seems really obvious and kind of makes sense, but I'd never, ever thought that there would be such a level of nuance at a national level around that.

And it really stuck with me.

I think especially with a pan European show like Shop Talk, Maybe we can't talk about it in exactly the same way because there is such a discrepancy between one country over another.

That's a really good point.

Yeah.

Rebecca, I'm still digesting and reflecting, so I'm gonna come back to you on that one.

Maybe just a separate one.

A separate.

Yeah.

Okay, we'll come back to you.

All right, Ben.

Okay.

My biggest epiphany.

I never knew how many New York Knicks fans there were in Talk Talk Europe until this morning.

Like, it's like the way that Arsenal fans have just appeared all over London.

Every third person is something Arsenal fan.

No.

So what's my biggest epiphany?

I mean when we've spoken a lot of the theory around LLMs, LM discovery and agentic, there's been almost this belief that this is going to be something pure.

It's going to be, it's a democratizing thing.

Anybody can be discovered if their products good enough.

And then the influencing is all about people talking about you, influencers, media, this whole new way to be discovered that there's going to be incredibly democratic.

We're already starting to see how the advertising gets plugged in.

So there is.

I keep getting picked up on this concern and this way it comes back to the previous point about understanding the algorithms, how do we get discovered and how does it change?

So one thing, one I spoke to a grocery CPG who's looked at how they get their brand discovered in the UK and it's pulling from some of the things you might have read about from Reddit, it's pulling from press releases.

They did exactly the same test with a vendor on the same same platforms in the Netherlands in Dutch and it all pulled from the out behind website.

So even a different market and a different language, your LLM sourcing is completely different.

So the complexity now of that strategy so that there is then the next level.

Okay, well how's advertising going to that?

Because I know how to get found on Google, I don't know how to get found on Amazon, I've just got to pay for it.

So how do we get found and what's his advertising going to be?

And then there was this extra build and this was the epiphany we've spoken about for a retailer.

What's the best way forward?

Do you want to create these apps within chat ggp?

Do you want to use Chat GPT or Google to be or do you want to build in your own website, your own apps and build your conversational commerce capabilities?

I had two brand owners say to me yeah, but will we trust the retailers if, if they do that?

And I hadn't thought about that at all.

So is that going to be another way and a whole new algorithm to learn and another sourcing model And I think, who knows that's all new and there's a lot to think about when it comes back to mastering the algorithms.

In that will they trust the retailers?

If the retailers do that, would it.

Still be as pure and democratic as the early view of AgentIQ?

Why do you think they feel that.

Way, I think because they used to trade spend.

Yeah, so we used to trade spend.

We know how that works.

Dynamics of the relationship.

It's the dynamics of the relationship.

Wow.

Rebecca, anything you want to add?

Yeah, so I've been spending quite a bit of time thinking about, well, thinking about tech, retail, tech innovation through the lens of VCs.

So you know, where's the funding going?

What are they seeing as defensible and investable solutions, technology companies.

And I think the biggest thing for me is, I mean you've heard in the media, scaremongering, SAS apocalypse, etc, so whether you agree with that or not is another conversation I think.

But hearing from the investors that we had on our panel, they are shifting kind of investment towards like hardware, still AI powered solutions but really focused on hardware, robotics to be using high fulfillment supply chain.

And their theory thesis basically is they have a bigger, larger moat around them.

They're more defensible, less likely I suppose to be copied by, well, less.

What's the word?

More defensible and less trying to think of the word more defensible.

Let's leave it there.

Or disrupted.

And I think that's interesting.

And also kind of we had two of the VC investors on the panel whose portfolios invest in both retail tech and consumer brands.

So when they were talking about the retail tech, they were really talking about hardware being really attractive to them and they were also talking about shifting investment towards more consumer brands as well.

Given the defensibility piece that we're talking about as well, I think that's quite a big epiphany.

It's really making me think about what the future of retail tech is going to look like and where the innovation kind of spots are going to be coming from, what retailers and brands, kind of the opportunities for them for partnering with tech companies, kind of what's going to come out of this new wave of investment or rather shift from software companies to robotic kind of hardware companies.

I think it's interesting, it's a massive change, just absolutely massive.

You know, mine is, mine is.

We kind of peppered the edges on it.

But Ben, it goes a little bit to what you said, democratizing.

I feel, I feel like that is potentially what AI will do.

Particularly the word I'd use is around cachet.

I think the definition of how we associate cache with different brands is going to change.

That's what I never thought about.

And Nadine Graff is the one that kind of tipped me off to this, which is before as retailers we would think about where do I allow my products to be placed.

I'm going to place them here in X, but not.

Yes.

And now when you hear, like, brands like Estee Lauder saying, you know what, we've just got to be on Amazon.

If we get to a world where all of the transactions are happening, you know, we saw different stats, 30% happening in the LLMs, whether agentic or not, a large portion will happen there.

How much does it just become then?

How quickly do we get them the products they want?

Which means the marketplaces, which are essentially E commerce, when you look at the volume of E commerce, it's all coming through my marketplaces.

Do more and more of the brands start to say, you know what, it just matters that we're available.

It doesn't matter.

The cache of where we're available becomes less and less important.

That's what I've been thinking a lot about.

Interesting.

All right, well, that wraps us up.

Ben, any final closing words here?

So, traditionally, I always thank everybody for attending, so I'm going to do exactly that.

Thank you for taking the time to be with us for the last three days.

We'll be back first to the third of June next year, but thank you all for coming and have really safe journeys home.

Thank you, everyone.

Yes, and as I say at the conclusion of every one of our podcasts, be careful out there.