Subscribe
Share
Share
Embed
Unlocking Retail Media is the essential podcast for leaders and marketers navigating the rapidly evolving world of retail advertising. We move beyond day-to-day operations to explore the strategic future of the industry, covering major investment trends, the shift to hybrid marketplace models, and the existential disruption posed by Agentic Commerce. Host James Avery brings in top industry veterans and visionary founders to analyze how ground-breaking technology is transforming customer journeys, influencing product catalogs, and forcing retailers to rethink on-site, in-store, and digital media strategies to remain competitive in the modern age.
Grant Steadman: [00:00:00] I'm one of, I guess, the AI maximalists. I'm more on that side of the scale than the, you know, the deniers on the other side of the scale. But I, I was pretty clear then that this is gonna change the customer interface of how they shop, um, and how c- and how they engage with retailers.
James Avery: Welcome to Unlocking Retail Media, the podcast where we explore the evolving world of retail media from data strategy to monetization and everything in between.
This is where we break down how retailers can build smarter data-driven media networks by aligning with what brands truly need from scalable ad solutions and meaningful metrics to cross-channel attribution and programmatic strategy. Welcome to Unlocking Retail Media. Today's guest is one of those people who's been working on retail media before it was called retail media.
Grant Stedman spent nearly 16 years at dunnhumby, the company behind the data science powering Tesco's Clubcard, Kroger's loyalty programs, and retail media networks across 22 countries. He rose to president [00:01:00] of North America, then went on to lead customer success at Coolerscreens, helping bring retail media into the physical store.
Today, he runs Reciprocal Advisory Ventures, co-founded Astroworks AI, and advises at Bain. Today, we're gonna dig into what building retail media actually looked like before it was hot, where retailers still get it wrong, and then we're gonna talk a lot about AI and get into how, how things are generally gonna change.
Grant, great to have you on.
Grant Steadman: James, fantastic to be here. Thanks for having me on.
James Avery: So I'd love to start, uh... I think, you know, we've got this like little bit of a trend on the podcast, I think, trying to, trying to bring people on that have been doing this for a long time, right? You know, retail media has been, you know, really hot in the last five years, but, you know, dunnhumby was kinda doing this 10 years ago, 15 years ago.
Give us a little bit about your background and what it was like in those early days.
Grant Steadman: Yeah. Uh, happy to. Um, I wish I could tell you I, you know, I saw this all coming and I knew, uh, way back then, uh, how big the, the industry would be. Um, I w- I started my career at, uh, above-the-line ad agencies as [00:02:00] one of the big Omnicom, uh, agencies in London.
I had some retail and CPG clients and was kinda dismayed at the lack of performance measurement or any kind of data on how these big campaigns that we'd invested so much in were performing. Um, so I started moving, you know, below the line. I went into direct marketing, dig- early days of digital. Uh, looking for data, looking for some answers beyond just, you know, how many we'd contacted or, or total sales.
Uh, I had Unilever as a, as a customer of mine, a client of mine, and they said that they'd started to work with this company called dunnhumby who were leveraging the, the, the data from what was then the new loyalty program that Tesco had launched. Um, so I, I found a role there, went out to move out to dunnhumby and, um, yeah, joined right into what they called their comms and media team.
It-- that went through various names. It was called the retail media, uh, team at one point. Um, it was very exciting. It was groundbreaking. Um, you know, dunnhumby was a very, still is a very data-centric company, so they had that-- they had started on that premise [00:03:00] of leveraging the data from the loyalty program.
I think that first, you know, a little before my time there, but I think it first went to personalization, so using the data to personalize the loyalty experience, the offers, the coupons, et cetera, and that was run out, um, out of dunnhumby. And then they were also monetizing the data more from a merchandising category management perspective.
And then it sort of more from there. So it's a, well, you know, dunnhumby are leveraging the data to plan and sell and, and measure personalized offers and, and also the data on what's going on in the categories. You know, Tesco obviously had a big shopper marketing, you know, that, that wasn't new. They had a big shopper marketing area.
They had in-store retail media, um, shopper marketing as it was, it was known as then, and they had magazines. So well, I think the conversation, I wasn't in the room with Tesco at that time, but I think the conversation was, "Well, why don't we bundle all those things together? You know, these guys are already using data for, for the offers and, and for the insights.
Why not bundle this all together and have them, you [00:04:00] know, use data to drive all those things in a, in a kind of holistic way?" So that's how it started. It was very exciting, fast-moving, you know, different organizations and reporting lines and names for the unit. But that's, uh-- So that was kind of late 2000s, you know, 2007, 2008, that kind of time.
James Avery: And was that, was that before-- 'cause dunnhumby now is a subsidiary of or part of Tesco. Was that-- When you started there, was it, was it already part of Tesco? It was just a, just a-- It
Grant Steadman: was already part-owned, and it was becoming a fully owned subsidiary, yeah, just after I joined, um, which was, I guess, a signal of how, you know, they had a big partnership with Tesco and how fundamental and- Yeah
strategic Tesco saw this opportunity. So that was a, you know, that was a, a, a, you know, a secret, uh, uh, that they, they obviously thought, "This is something we should fully own and, and leverage." Um, dunnhumby already had a, a global business as well, and I think it's 2003, I think they originally started working with Kroger.
Um, and some of these areas were big parts of the Kroger work at that time. [00:05:00] Um, others were, were less so. But, um, so yeah, they, they bought that out and then... You know, but, uh, retail media was, was one part of what they were doing, and it was an increasing part, but I wouldn't say it was a foregone conclusion that that was-- it was gonna turn into a huge in-industry and, you know, I was quite heads down and working on my campaigns and deliver- with my customers and delivering for those.
Um, and it felt like it was quite a unique capability for Dunnhumby, so it wasn't obvious, you know, that you would say, "Oh, this is something every retailer's gonna do globally." Um, there was a couple of guys there who, who broke off, um, to start up similar retail media businesses with other... I don't know if you know Matt Lee and, and Joel Hopwood.
They-- I worked with them a little bit in my early days at Dunnhumby, and they, I think around 2008, they broke off to do similar, uh, or, you know, retail media shopper marketing work with other chains there. So that was a clue. The results were always strong. The results- Yeah ... you know, come from an ad world where, you know, half of a percent or [00:06:00] point one of a percent was considered a good response rate.
So to see what-- how data could really transform the performance. So that was a clue, and it felt like some of these other banners in the UK. I think Alibaba launched digitally or its retail media, you know, ad sales division around that same sort of time. But it wasn't really, I guess, until like the early teens, twenty-- ten, twelve that it...
You know, I remember reading Amazon had set up Amazon Marketing Services, I think, I think it was called- Yeah ... at the time. And there was a couple of ex-Dunnhumby folks involved in that somewhere. But so that's when I knew. I thought, "This is gonna be really big now." And, um, and then of course, I think Roundel and Kroger, my former colleagues from eighty four fifty one, launched Kroger Precision Marketing around twenty sixteen, seventeen.
And I think by then Amazon was four billion in-
James Avery: Yeah, they were off to the, they were off to the races and everybody else- Yeah, it was a foregone conclusion ... was catching up. Yeah. And then who-- eighty four sixty one or eighty four fifty one, that, that was basically Dunnhumby at one [00:07:00] point, right? That kinda spun, spun over to Kroger?
Grant Steadman: Yeah. I mean, Dunnhumby had a j- a, a, a long-term partnership with Kroger from two thousand and three, I think for, for twelve years that ran. And I think Kroger made... You know, I wasn't close to it at the time. I was still, still over the pond. I was in the Nordics at that point. But, um, I think Kroger made this decision, it, it looks like from, from the, the outside to, to bring that in-house, and they effectively internalized and bought out a large part of what was Dunnhumby USA.
Dunnhumby USA had a few other customers or clients, retailers that it was working with. So I was part of the group that came over in the, in the couple of years after that to help build back up, you know, Dunnhumby outside of just purely the Kroger, uh, partnership. But yeah, l- all, a lot of eighty four fifty one, uh, folks were former colleagues, um, over here and, um, you know, took that on in, in their own way, leveraged some of the same tech and thought, the thinking and, you know, philosophies.
But, um, yeah, went in their own direction, obviously focused on Kroger.
James Avery: Yeah, it's funny over in, uh, in the UK it's, it's [00:08:00] kinda like everybody we talk to worked at Dunnhumby at one point. Yeah. Where the same way in the US it's like, it's almost like Amazon, right? There's just so many people that, like, came outta-- they did their tour at Amazon and then, and then came out to a retailer or work somewhere else.
Uh, Dunnhumby has that same effect in the UK. Uh, and you've had-- but you've kinda seen both sides, right? 'Cause, uh, you know, you're in the Nordics, you're in UK, over here in the US. Like, what do you, what do you see-- what do you think the main differences are? What do you see different about how the, how retail media's kind of evolved, uh, in Europe versus over here?
Grant Steadman: Yeah. Very different, like the scale I think for one thing is, is very different. Um, but also the context of, of the markets and how they've evolved I think is important and, and often gets skipped over. Um, I would say US is, you know, more commercially aggressive, more tech-forward, more innovative. They'll-- you know, retail media kind of started to some degree with a large retailer marketplace in Amazon.
Bigger ad budgets, m- invest more in tech, look to evolve [00:09:00] more and quicker. Um, so the retail media sales growth has been faster. It was more digital first and digi-digitally led, given that starting point and the scale of that, versus, you know, the UK had been more, um... And, and most of the European markets had started more in-store and shopper marketing, and then digital came along- Yeah
so kinda different context there. Um, UK and EU, you know, I think more loyalty and insight led, more aligned commercially within the retailer, more joined up from an omni-channel perspective. Like, you know, generally speaking. There's always- Right ... exceptions, of course. You know, Tesco was kind of the reference model in the UK, as we talked about with Dunnhumby.
Um, so I think that sort of stronger foundation and discipline around aligning retail media with what you're doing for customers overall and what you're doing from a merchandising perspective and a-- how you're trying to grow the category and more omni-channel. Uh, but, but just less scale, and I, and probably less, you know, less tech innovation.
And then you've got smaller European markets like, you know, my [00:10:00] friends in, in, in Norway, for example, um, won't mind me saying that, where, you know, things are a bit more pragmatic and you've got, you know, two or three large-scale retailers. You've got- Right ... very tight and very o-often local supplier communities who don't have big global, you know, brand budgets to tap into.
So that tends to be more constr- even more constrained by scale. Data tends to be more fragment-fragmented, less, less tech, um, you know, investment and, and a lo- real varying tech capabilities from retailer to retailer, market to market there. Less kind of, even less standardization, um, say in the Nordics compared to the US.
James Avery: Awesome. This is-- I love, I kinda love the, the history and background of, of retail media. Uh, but I think when we, we were hanging out at, I think it was NRF, uh, earlier this year, and- And it was really interesting that it looks like you've, you've kind of really shifted into thinking about AI and how AI's affecting retailers and retail media, and I think that was, that was the [00:11:00] majority of the conversation then.
So I think it's something everybody... It's top of mind for everybody. So I'd love to dig into, you know, let's, let's, let's talk about, uh, you know, what's, what's going on in the world of AI when it applies to, you know, retailers and retail media.
Grant Steadman: Yeah. Well, you know, I, I broke off on my own, uh, about three, three years and, and a- and change ago now, and, um, first thing I did was go to study AI at UC Berkeley, which was a good, uh, good decision and investment and, and learned a ton.
And, uh, you know, pretty quickly saw this was not just another data tool or a, or a tech, you know, solution. Um, you know, my perspective, and I'm one of I guess the AI maximalists. I'm more on that side of the scale than the, you know, the deniers on the other side of the scale. But I, I was pretty clear then that this is gonna change the customer interface of how they shop, um, and how c- and how they engage with retailers.
This is gonna change also the internal operating systems and model of [00:12:00] retailers within their businesses, and therefore it's gonna change the economics of, of, of retail. So, um, you know, I think that, that was my belief then, um, for re- and, and that's playing out. I'd say it's still kinda early days. We're probably two years into a, you know, another decade-plus wave of, of tech change.
Um, and, and so, you know, we're not remotely, you know, through it yet or, or really into it. We're probably only 10, 20% of the way, the way into the full value and, and potential of it. For retail media specifically, I think AI is gonna change how audiences are built, how campaigns are planned, how creative is versioned and automated and adapted and tested, how you optimize performance.
H- And, and from an agentic or AI-assisted perspective, you know, how shoppers discover, decide, um, engage with, with different products. So there's plenty of hype about it, of course, um, but there's real change already and, and, you know, from a planning and targeting and [00:13:00] creative versioning perspective. Also a, a little bit from measurement.
You know, AI is already, you know, penetrating that and, and improving that. Um- Yeah ... I'd say-
James Avery: Yeah, I kinda think, I think I was, like, breaking it into, like, if we look at it in a couple different sections where there's kind of the, there's the, the piece you were just talking about, which is kind of how has AI improved, like, how an RMN operates, right?
Whether it's better measurement, better creative, better targeting. But I think first let's go to the other side of it, which I think you were, where you first started talking about, which is, like, how do you think... You know, this is, like, the big question about, you know, agentic commerce and, and how, how does, how does it change, like, buying behavior, right?
Like, how do you... You know, do you, do you, do you vision... I feel like the, the maximalist standpoint is kinda like, hey, like- We're all just gonna buy things talking to an agent, right? Like, if I'm, if I'm looking for a new polo shirt, I'm just gonna have a conversation and it's gonna talk about a d- couple different brands.
I'm gonna say, "Yes, send that to my door." Uh, whereas I say the, the, the [00:14:00] flip side of that conversation or the argument is, you know, no, people are still gonna shop, right? Like, e-commerce is only 25% of, of commerce, right? Like, there's, you know- Yeah ... it's, there's still-- people are still gonna go feel a shirt and try it on and, and comparison shop and, and kind of where...
How do you think... Where do you think we are in, you know, say a couple years, you know, versus, say 10 years?
Grant Steadman: Yeah. I think, you know, it's still nascent today, so I think that it's a time horizon question, and it's a relative scale question, right? So I mean, a lot of people said this about e-commerce, "Oh, the stores are dead.
No one's gonna shop the stores." It's still, for grocery, it's still 85, 90% of sales in the store. So, uh, it's almost worse than that. And, you know, I was talking to someone about this earlier today. It's not like agentic commerce or AI-assisted commerce is a fully automated shopping experience that's gonna replace everything.
Um- Yeah ... it, it's, it's a whole nother bunch of things. You know, if you wanna meet the customer where they are as a retailer and a brand, if you wanna be present where the customer is, you'll need to have physical stores, and you'll need to have a website, [00:15:00] and that will need to have functionality, a lot of the functionality it has today, and you will need to show up in AI agents as well as having your own integrations in those AI agents.
So I think it's just, it's not a, you know, all those things, b- both, um, multiple things can be true at the same time. Um, I do, you know, I think we're already seeing it sh- show up in discovery and search, right? The customers are already searching, and it varies on different categories and different demographics of customers and what have you.
But generally, there's a, a very broad audience using AI assistance, and they're already using it for AI d- for, for discovery of different products, for searching for different products. Um, and then some of those go on to the next phase, and they, you know, complete a purchase or part of a purchase, or they assemble a basket through an AI platform, and others don't.
Others take that list, and they go into the store with it, or they, you know, go to the retailer website. So it's a, you know, it's a, it's a fragment- it's gonna be a fragmented journey. It's not gonna go from everyone, you know, shopping in one way to everyone shopping everything through- Right ... through ChatGPT.
It's
James Avery: kinda, [00:16:00] it's, it's almost more like, um, like mobile, right? Where it was like e-commerce, e-commerce was already a thing, right? Um- Yeah ... and then mobile comes along, and now you need to have an app. Users wanna be able to buy through the app, you know, and maybe they were buying on the website, and it's a, it's another channel.
It's another extension of e-commerce where, you know, agentic kind of just feels like another, it's another app store you kind of have to be in. Uh, and it's just gonna be a, a, a third way, right? That, that people are kind of interacting with your, your e-commerce portal.
Grant Steadman: Right. And I think that's true, but I think also how customers shop all those channels is changing.
So they are becoming used to AI-assisted interfaces in, in all sorts of use cases, but shopping and search and discovery and execution and, you know, the actual transaction is also one of them. But they're, but they're getting used to AI helping them in a more intelligent way and helping them get to solutions in a quicker, faster, better way.[00:17:00]
And, you know, I... There's a lot of... So I think in, in all of those channels that c- customers w- will, will be used and expect those kind of AI-assisted experiences, whether that's through the whole funnel through to completing of the transaction and the, and the payment, it... or, or just parts of it is, is another question, right?
There, there will be that. But w- where it's not all the, all automated. But- Yeah ... you got AI-assisted elements of the journey are gonna be needed on retailers' on-site websites as well as, you know, on these AI platforms.
James Avery: Yeah. So I was gonna, I was gonna, I was gonna ask you about that because I think there's, there's kind of when we think about the AI shopping experience, there's, there's the new channel, which would be like, okay, you're connecting into ChatGPT.
I can now see your products, maybe even transact through ChatGPT to buy a product. That's kind of a whole new channel. And then there's the, you know, there's the, the agents on site, right? Like Amazons or Targets, like... And I think we- we're starting to see that more and more of, okay, yeah, you go to the site, you go to the [00:18:00] mobile app, you wanna, you wanna engage with a, with an agentic shopping experience on that retailer.
Grant Steadman: Right. And so it's
James Avery: almost like a- Yeah ... like you said, right? It's a, it's an AI enablement of all of those existing channels, but it's also kind of a whole new channel if we're thinking about the, the, you know, ChatGPT or Claude or, or Perplexity or
Grant Steadman: whatever. Yeah. So there's a bunch of new channels there, and then there's also a new, new way to access and new, new different kinds of solutions on the existing channels as well, like on, on the, the retailer's existing websites.
And I think it's a real... You know, there's a lot of sort of talk of, oh, this is a big threat to, to, to retail media, you know. And, and I do think that from a traffic and clicks perspective, there will impact there. The question is, can AI-assisted shopping, can retail media support that in a different way?
And are there opportunities? You know, there are threats to the current models and the current activations and how that looks, but I really think it's an opportunity actually to have a much more, for retail media support, a much more intelligent, helpful [00:19:00] conversation with the customer. If you think about re- a retail media interaction today, if that was a conversation with a customer, it would be customer going, "Ketchup" in the- Yeah, yeah
search bar, and, and the, the retailer or the brand going back, "We have Heinz this size or so this... You know, the, the, not to get into the brands, but, you know, it's a very basic, simple conversation which then involves a lot of manual work f-from there to, you know, a customer g-explaining their problem, explaining their challenge.
"I'm having a dinner party," "I'm having a picnic for 12 people, two are vegans, two are this," et cetera. "How do we..." And then you building a whole base and really getting- Yeah ... quicker and better to a better solution for them. I think retail media can help in that journey, but it just looks very different than the, the static, you know, search and, and keyword-based approach today, for example.
James Avery: Yeah. And basically it's, it's just like, I mean, going all the way back to, tying it back to Dunnhumby and data, right? It's, it's just like using that data to inform better promotions and better offers. Right. The user's [00:20:00] now just telling you more, right? Like, it used to be you had to, you had to know, okay, Grant, Grant buys this type of ketchup and he buys it every other week with his club card, and so we're gonna target him this way, to where now you might come to the site and say, "You know, I'm having a cookout."
Right. "I'm gonna need, I'm gonna need a lot of ketchup," right? Like, oh, well, here you go. Here's the 64-ounce ketchup, you know, like, uh, versus what you might normally buy, right? It's like it's, you're actually getting a lot more intent, like a lot stronger intent signal and information from the customer, but it's up to us as the, you know, the platforms to actually use that data- Right
as opposed to just see ketchup and say, "Here's some ketchup." Right.
Grant Steadman: Yeah, and I think there is a strategic kind of control issue for retailers about, you know, they've owned the customer, they've owned... They've been the primary interface receiving that, that transaction data, that loyalty data in the past.
And so are they gonna maintain that if, if that goes through ChatGPT and the customer puts a prompt in there? Like, who owns that signal and how is that signal shared? But to your point, those are really rich new signals [00:21:00] and all kinds of context about the customers and their lives and what they're looking for that if those can be harnessed and used really intelligently, I think it can, you know, it can support retail media and just in a, in a very different way.
Um, you know, I suppose there's a risk of it more fragmentation and some people have this type of data signal and don't have that one, and it, it all gets a bit more patchwork. But, um, I think the retailers need to decide are they, you know, how are they gonna maintain that data ownership and control, who's gonna own the intelligence layer, um, the, the p- the data powers and that is used to drive those things.
But I, I think that will be sol- that's solvable, and I, I'm, you know, I'm positive on it. I, I, I'm not saying the g- the growth rate of retail media outside of other channels if you, you know, exclude agentic commerce, I'm not saying that will continue necessarily. Um, but I think as retail media as a whole, if it embraces these These new interfaces, these new experiences, it can, you know, it can really help it, [00:22:00] um, maintain and grow.
James Avery: Yeah, I like, I like to say, uh, advertising will find a way. Uh- It- We always, we, you know, it's like Jurassic Park. We will, we will find, we will find a way to integrate into any new format, any new application, any, any interface. Uh, we will, we will find a way to get in there.
Grant Steadman: Well, and it's inevitable because the, it has to be.
I mean, the, the cost of, you know, if you look at the AI investments on the, on the hardware and the, the chips and everything, that's, that's got to ... Something's got to pay for that, and advertising is a, is a good candidate. So it, th- they, you know, there's a commercial necessity to find a way- Yeah ... to do that, and I think, I think we will.
And I think it's be good, I think it'd be good for everyone, but it's gonna be change, and some people don't like that, and some people get very vested in their current economic interests and operating model and don't wanna see around that, and so they kind of deny the, the, the tech- the change is coming.
But, um, uh, I think it's, yeah. It may- maybe, maybe a couple more years than some people predict. I've seen a lot of predictions for sort of 2030, it's gonna be this much of, you know, [00:23:00] 30% of all e-commerce is gonna be through AI agents, and it all depends on how you define that. I'm not sure it's gonna be quite that aggressive in, in the growth, but I do, I do think it's coming and inevitable.
James Avery: Yeah. Well, and the funny thing is if you're, if you're 30%, 30% of all key mar- i- if 30% of all e-commerce is agentic, you're still talking about 10% of all commerce. Uh, and so thinking about that, uh, if you're a retailer, what are, you know, where ... You know, they're hearing all of these things. They've got, you know, got ChatGPT over here.
You've got, "Hey, you guys should have your own agent. Hey, you guys should be thinking about how your, how your LLMs are looking at your catalog." Like, where ... If, if you're talking to a retailer, like, what, how do you, how do you suggest they attack this opportunity?
Grant Steadman: Well, I think there's certain, um ... You know, we talk about of, and this is some of the work we do, um, at Astroworx, you know, helping retailers on their AI transformations.
We talk about four levels aro- around the sort of s- strategy, your activations, your change management, and your enablers. Um, and, you know, the enablers being kind of the data and the [00:24:00] models and the governance and, and some, you know, kind of basic training and what have you. So we talk about those different levels.
I would say, you know, what is, what is least ... And a lot of that stuff, you know, "Oh, yes, I know we need this, I know we need this," and, you know, and, and they're aware of that. What is most understood, I think, most misunderstood and skipped over is, you know, th- people tend to dive into the tech, and it's like, "Okay, this is a tech challenge, and we need to approach this like we did digital transformation, and we need some new tech, and we need some new data."
And that's, that's all true and part of it, but what- That's actually for me, and what we've seen, is an easier solve than the leadership and the organizational challenge that it represents. You know, it's, it's really an operating challenge to embed these things, th-this, this AI, and make it work across multiple use cases.
And retail media is one area. But it's, it's really an, an operational and organizational challenge to embed it and change how people work to embrace it and get the most out of it. Um, it's not just an automated and productivity challenge. It's... You know, [00:25:00] that's one kind of use case, but that's too narrow.
You know, the, the biggest change is that AI can change how decisions are made, it changes the jobs people do. So we do a lot of work on, you know, the change management around what is the workflow today in, in this particular process or function? How is AI gonna change that? And then what-- So what does the, the tomorrow's process look like?
And, and sometimes, you know, and there's different types of AI that have different impacts on that. Sometimes it's a very similar process, just much faster, uh, and speed is a big, you know, advantage of AI. Other times you're cutting across multiple functions and teams, that you're, you're changing decision rights, you're changing who does what.
It's a big kind of race-y, uh, adaptation piece of work and, and- Yeah ... stakeholders need managing. So that I think is often o-overlooked, um, and, and underestimated. And then i-i-in, in those cases, the retailers run into pilots, but then they have struggle scaling them because, uh, uh, you know, doing a one-off kind of s-side desk exercise is one [00:26:00] thing, but making that the default way of working across, cutting across functions is, is a much bigger change process.
So that's, um, often, you know, overlooked, I would say.
James Avery: Yeah. And Heidi, I know, uh, you know, I think we're, we're recording this a little bit before it comes out, and I promise we'll, we'll hold it until, until you release. Uh, I know you have a, you have a report coming out soon. Do you wanna talk a little bit about, about the report and kind of maybe some of the highlights, so people can go download it and, and, and check it out after, after listening?
Grant Steadman: Yeah. Thank you, James. I'd, I'll be pleased to. Um, so yeah, I mentioned Astroworks and, and that the company and what we're doing there on AI transformation. So we are, we are publishing the, the what we believe is the first study in, in retail, um, on how retailers are deploying AI-assisted or agentic commerce capabilities, um, for their customers.
Um, so it's called the Agentic Commerce, uh, Capability Index, um, which it comes out on, uh, May twenty-first, so by the time this is out, I think people will be able to access it. Um, and it's, it's focused on twenty-four of the [00:27:00] leading grocery retailers, and we-we're starting in that, in that sector. Um, so it's still in the field.
We've been doing a lot of different rounds of research. It's been challenging 'cause these capabilities are evolving every week, and I, you know, the what keeps me awake at night is, uh, a big, a big new launch from a retailer on an agentic commerce capability in the middle of our final, uh, field study window.
Um, but effectively it's trying to say, you know, what do retailers, um, deliver for their customers today in terms of AI assistance, both, um, and there's different dimensions to it, um, both on-site, so on their own digital properties. Is there an AI assistant on their website or their app, and how, what functionality does that have?
And we drill right down and run different prompts and scenarios to kind of understand how that works. Um, similarly, off-site, you know, do, do these, which retailers have integrations with ChatGPT and Claude and Perplexity and Gemini, and what do those look like? And what do they look like if through a plugin or an integration versus just an open organic chat experience?
Um, so drilling into that. [00:28:00] Um, we're also looking at, you know, commerce media and, um, and how that, and, and, you know, is, is media being tested and retail media being tested on those AI interfaces? And very early days in that, it's really hard to capture, um, any live examples 'cause of the, the way it's been tested.
But looking at which retailers are embracing that and to what extent and, and how, how that's working. Um, and also looking at some of the data elements of what does, what, what signals can we get from how these experiences work about how they're investing, um, from a, you know, data and technology perspective in, in supporting these foundations.
So that's the study, and then basically the findings show that there's a handful, and I, you know, I won't go too much into, into it in case things change at the last minute. But, you know, as you can imagine, the likes of Amazon and Walmart leading the way with their capabilities. Amazon particularly on its own, own properties and what it's doing through Rufus, um, and Walmart of course through Sparky, uh, similarly, but also on the AI platforms.
Um, they're really leading the [00:29:00] way at the moment with a lot of functionality, and that's reflective of their overall strategy. You've then got a follower group and, you know, the Kroger company and Albertsons and a few others are, are, are in that space where they're doing a few things. They seem to be testing.
They don't yet seem to be at full scale or, or going all in with any one, um, AI platform. And then, then you've got a, you know, a long tail of folks who are maybe doing one or two things. Um, and then you've got a, you know, a couple of retailers who are mainstream, you know, folks like Trader Joe's and WinCo and typically value-oriented chains who just aren't playing in the space really at all and blocking it and, you know, not, not being picked up there, which is, you know, a, a, a strategic choice they're making.
Um- Yeah.
James Avery: It's a choice. Yeah. Y-
Grant Steadman: yeah, it's a choice. Um, you know, we'll see how, how long it lasts. But, um, so that's how it looks today, and our idea is to, you know, we will refresh this on a regular basis and keep this research going and see how, which retailers are evolving and, and how these different elements of the, the experience, um, becomes more [00:30:00] important over time.
James Avery: And let's-- So if you're... And I think this is interesting. If you're-- Let's say you're, you're a grocer kind of further down the list, like maybe you really haven't, you haven't done a lot yet. Where, where would you, where would you... I guess maybe it's, you know, depends on a lot of questions. But, but thinking about like, is it, is it leaning into, you know, how am I exposed in the LLMs?
Or is it, is it trying to keep that consumer on your site, right? To keep them in your own agentic, you know, in your version of Sparky or Rufus, right? Like, it's kinda like if I, if I'm a retailer, like where do, where do you, where do you think about, how do you think about placing that bet?
Grant Steadman: I mean, if I'm a retailer, I, I, it's a strategic choice.
I think it's gotta reflect your overall strategy and your overall positioning. Like I said, some of the price-driving, you know, price-driven, store-heavy, you know, th-they've been slow on e-commerce, and they've not really invested in that as a decision, and I think it's, it's similar here. So it's gotta reflect your overall brand and value proposition.
Um, but, you know, my belief is if you're gonna meet the customers where they are, they are gonna be Uh, they are shopping already on [00:31:00] ch- on these AI platforms- Yeah ... looking, looking for products. So you need to show up there. And if they're gonna go to your website and they're going to, you know, go to the search bar, and pretty soon they'll expect that if they put in a long prompt telling you all about their summer barbecues and all the, that you, you give more of a response than just, "We sell ketchup," "This ketchup," right?
Yeah, yeah, yeah. So you need to have those cap- I believe you will need to have those capabilities. So some will not do that, some at all ever, some will be slow to it. Others are already kind of going down that road. So this is trying to benchmark, like, how are, how are retailers thinking about that? What are they offering?
How that... And, you know, we'll track how that kind of evolves over time. But I, I think you have to get there, and that's what I'm saying. I think it's, uh, it's not just the platforms themselves, it's the way customers shop is gonna change. They, they will expect to get more intelligence back and more solutions back and, and less just manual product-by-product keyword- Yeah
type thing. So I think you have to meet them there. But some retailers won't, and some of them will still do really well because they're great value, [00:32:00] they've got great locations. The s- the start... You know, you don't have to compete on... And, and some of the smaller chains can't afford to compete. Um, the smaller regional supermarkets, you know, a lot of them work with Instacart.
Instacart have some fantastic AI-assisted capabilities, and, and these guys will, will benefit from that, being part of that ecosystem. Right. I would a- I would also want to show up as my own brand as a retailer in those platforms and not just be reliant on, on an Instacart integration. I would, you know, I would build a plugin and an in- you know, an integration as, as Target and, and Walmart and, um, and other banners have.
So when you go to ChatGPT, you have a sort of branded environment that fully reflects you as a retailer, that your customers can find you there. Um, but, you know, it's, it's a, for some of the smaller folks it's gonna be a bandwidth discussion. They can't do everything.
James Avery: Yeah. It's kind of an interesting opportunity, though, if you're, if you're a brand, if you're a retailer who, who kind of was late to e-commerce and, and didn't build up, you know, a big, you know, e-commerce [00:33:00] operation.
It's kind of an interesting opportunity to say, like, if you, if you really went in on it, you could, you could kind of just, you know, leapfrog a lot of people who did, right? By really, really jumping in on this and saying, "Hey, we're not, we're not gonna go..." You know, maybe they never built a really powerful website and search engine and, you know, they're like, "Hey, we're just gonna, we're just gonna start AI native," uh- Right
as opposed to, as opposed to kind of going through the last 20 years that everybody else went through.
Grant Steadman: Exactly. And I think, you know, AI, the promise of AI should be a leveler of the playing field for smaller chains who, who wanna make that strategic decision that they can do that leapfrog. You don't need hundreds of analysts, hundreds of data scientists.
Yeah. You don't need big infrastructure that only the largest national chains have had. You know, you don't need that. You can u- if you leverage AI properly, you can build these things, but it relies on you, you know, going back to the change management and organizational challenge, it relies on you being more agile and more flexible and adjusting how you work.
But if you can do that, and you can leverage AI, then I think there's a real leapfrog [00:34:00] opportunity there. But you gotta be, you gotta be set up operationally to, to deliver on that, and, um, you know, that's, that's tough for a lot of retailers with all the day-to-day, you know, operational challenges and, and work to do.
James Avery: Yeah. Well, awesome. Grant, this has been, it's been great to have you on. I think this has been a great conversation. I hope, uh, we'll, we'll, we'll put a link to the, to the report where people can find it kinda in the, in the show notes and the LinkedIn post. Uh, but I'm sure we'll probably, probably have you back in the future where, you know, in, probably in six months' time we'll have, uh, have had three years of, uh, AI time.
So it'll be a lot for us to, to cover again.
Grant Steadman: No, I'd love to come back, and, uh, yeah, it's always a pleasure talking to you, James. Thanks for, thanks for having me on. Awesome. Thank you.
James Avery: Thanks for tuning in to Unlocking Retail Media. If you enjoyed this episode, don't forget to subscribe and share this show with your network.
We'll be back soon with more insights to help you navigate the future of retail media. See you next time.