# S3 E8 T63 - Mark Williamson - Costco II

S3 E8 T63 - Mark Williamson - Costco II 

[00:00:00] Tom: Welcome to the Middlemen podcast. We are here again. This is the first time we've had a guest on twice and couldn't be a more exciting person to have. Mark Williamson, uh, who does retail media at Costco. Welcome, mark. 

[00:00:17] Mark Williamson: Hey, how's it going? 

[00:00:19] Tom: Really good. Uh, I have some good news and some bad news for you. Bad news. First is you are no longer our top episode on the Middlemen podcast. The good news is you're back so you can rela reclaim your title. 

[00:00:30] Mark Williamson: Do you, uh, do you aggregate listens between two episodes, or do I have to start from scratch? 

[00:00:36] Tom: That is a good question because we have not had anybody on. I will, I will defer that answer, but, uh, but we're, we can, we can figure out a way for you to become number one again. 

[00:00:45] Mark Williamson: All 

[00:00:46] Tom: Um. 

[00:00:46] Mark Williamson: I always appreciate the challenge. 

[00:00:48] Tom: Maybe, um, we should say that you are now an author of a new LinkedIn newsletter, the Buck 50 Newsletter, which I'm a proud subscriber of. 

[00:00:57] Tom: And you know, it's funny, when we [00:01:00] first, uh, when we started out the year I saw you at NRF, the Strata Cash Event, great event. Um, and Kerry Masters talked about, you know, ad nauseum, your, your, your tech stack and the fact that you opened the kimono and that was great. Um, but what, honestly, there was a lot in there that I wasn't sure about. 

[00:01:18] Tom: There was a lot of new vendors, so it took me some time to process. But when I read your Buck 50 Newsletter episode or installment two about Google product listing ads, I was like, ah, something totally new. How did you get to running Google product listing? 

[00:01:34] Mark Williamson: Um, yeah, it's a great question. Uh, we, uh, it, it's something that we've wanted to do for a while, uh, mostly because we, we see it as an opportunity. Uh, we get asked about it, uh, I would say fairly regularly from advertising partners. Uh, and I think it's because, uh, our, uh, our e-commerce business is. Is so centered on what we call big and bulky. 

[00:01:58] Mark Williamson: And so the items that [00:02:00] we sell the most of that have high e-com penetration for us also happen to be items that are regularly searched. And so there is this kind of, I would say, this business imperative to say, um, like are, like to ask ourselves, are we showing up? Are the items. Uh, our, our supplier items showing up wherever people are searching. 

[00:02:20] Mark Williamson: And I like the premise of this, this, not just this channel, but of, of everything that we're doing is like, we don't have the right to tell people how to search or how to shop or how to research or whatever it is they're doing. They're gonna do what they're gonna do. And, and it's our responsibility to be present wherever searches are happening. 

[00:02:37] Mark Williamson: And that's, that is, it's changing in real time. It's gonna continue to change. And so we really just challenged ourselves to say where, where do we need to show up? And for, for PLA in particular, like the timing of this really doesn't have much to do with anything other than the fact that we were technically ready to do it. 

[00:02:53] Mark Williamson: And so we've been doing a lot of stuff on our kind of foundational infrastructure, uh, and that that's really what I talked about at NRF [00:03:00] is this idea of having a, a holistic retail media platform that is built on our first party data that is aligned with our consent and preference and our identity management. 

[00:03:10] Mark Williamson: So we can move into new channels, including upstream channels, to be able to, uh, take, uh, the capabilities that we have as a closed loop retail media offering into, uh, into new spaces. And so we just, so, like we've been have, we've had plans with symbiosis for a long time. We've had plans for this channel for a long time. 

[00:03:30] Mark Williamson: We were literally just waiting for the infrastructure to be ready for, uh, for this channel to come to life. 

[00:03:36] Tom: so to back that up, for the people who weren't in NRF, you're talking about sort of meta router, transcend growth loop. Are all those part of this? Is that necessary to make this work? 

[00:03:47] Mark Williamson: Uh, yeah, one, one of those, well, actually two of those incredibly important. So with, uh, with Transcend in terms of like, that's how we manage, uh, privacy compliance and to making, making sure that any data that [00:04:00] we're utilizing or capturing. Warehousing, all of that is as compliant as it needs to be. Uh, and hits our very, very high benchmark of, of, uh, maintaining and, and hopefully growing member trust through all of this. 

[00:04:13] Mark Williamson: And so, so yeah, it transcends very much a component of this. And then meta router is, for all intents and purposes, our, our data routing. Uh, ecosystem, which, because there's signals that we create, there's signals that other partner channels create and matter outer is what facilitates the orchestration of that. 

[00:04:31] Mark Williamson: The data flows between, uh, between the whole, uh, the whole ecosystem. And so in order to capture. Uh, the signals from offsite demand. We leverage red, uh, meta router to, uh, to help figure out who that is, uh, what they've done, and then, uh, and then feeds that information, those signals back into our, uh, our unified data platform. 

[00:04:53] Mark Williamson: Uh, so we can do the close of measurement, so we can look at things like remarketing. So those two are, are very critical. Um, growth Loop is [00:05:00] a, is a, uh, uh, we're using as an audience builder today. And so they, it, it has less of an impact, uh, with, uh, with this particular channel, but the signals that we capture from this channel, like those in-market, high intent signals that feed into our unified data platform then become usable within an audience. 

[00:05:20] Mark Williamson: And so, uh, growth Loop is not very, very far away from this channel at all. 

[00:05:25] Todd: You've been doing offsite, previous to this, so this is, that's not necessarily thing, it's the, in terms of Google and PLA is initiative Here, 

[00:05:38] Tom: Todd, your, your mic isn't coming through very well. 

[00:05:43] Todd: we can, let me switch. I. 

[00:05:47] Tom: I will just not say anything so that there's no echo. Yep. 

[00:05:52] Todd: is that better? Um, what I was saying is that the, you, uh, mark [00:06:00] offsite is something you guys have been, you did previous to the Google function. So that's not, the new news is offsite. What is new is using Google at. Channel. Um, I, what's interesting is that when we think about offsite, I'm not sure many other retailers have used PLA as much or go shopping as much for offsite. 

[00:06:19] Todd: So it's interesting to hear about that, I think as another activation channel for off. 

[00:06:27] Mark Williamson: Yeah, you're right. We, we've been doing, uh, offsite, programmatic for over a year now, and so that's, that's really where we started that journey. And so the, the premise is basically the same. Is, is that. Our members are consuming content. They're researching, they're, they're doing things outside of our ecosystem, and so we have to figure out how to engage them, uh, in, in the right way. 

[00:06:50] Mark Williamson: And so to me, like Google, PLA is is just another format. Uh, and, and so it's, it's not a display ad. It's not [00:07:00] connected tv, but it's still a place that something is happening. It's, it's lower down the funnel. It's definitely mid funnel. It's not necessarily, you know, it's very different in purpose than a connected TV ad. 

[00:07:10] Mark Williamson: But foundationally we need to be able to do the same things, whether it's targeting, uh, signal capture. Uh, uh, measurement, all of those things have to be the same. And so for us, this is just an extension of that offsite pillar that, that solves a unique need or solves for a certain channel, but doesn't sacrifice any of the capabilities that we expect to have in the kind of open internet programmatic space. 

[00:07:34] Mark Williamson: So Google as a, as a walled garden functions differently. And so that's why it was, it's very critical for us to get our ducks in a row so we can then interface with them, so we can be interoperable with them. In the way that we want, uh, we want things to happen because our advertisers are demanding, targeting in-flight optimization based on real conversion signals, measurement, these sorts of things. 

[00:07:56] Mark Williamson: And so, like, in some extent we are going channel by channel, and [00:08:00] the open Internet's a little bit easier. Programmatic is a little bit easier. It's harder to nail down some of the walled gardens, but what we're finding is that with the right focus, we can get that part done. And so, um, so yeah, we, we definitely want to cover the full funnel. 

[00:08:13] Mark Williamson: Top to bottom. Uh, but we don't want to make any sacrifices in any of the channels. So the promise of what Costco can do as a closed loop partner is true at all aspects. Even if behind the scenes the wiring looks a little bit different. Uh, channel by channel, just by necessity. 

[00:08:31] Tom: So what is the, you know, you mentioned symbiosis, which is the partner that's recently acquired by DoorDash, but that's, that's powering this for you. What is it that they do specifically, like talk? Talk a little bit about how they're a vendor in between you and say Google, because what I'd heard of of symbiosis is that they're doing things like collaborative bidding, which sounds really interesting, but can you just sort of like give us a context there? 

[00:08:56] Mark Williamson: Yeah. Uh, yeah. So symbiosis, uh, like, [00:09:00] like at its at a foundational level, we need a way to manage campaigns, to make buys from Google. And then to be able to optimize and measure and that sort of thing. And so like at its very core foundation symbiosis is we're being, is being leveraged by us as a campaign management and measurement tool. 

[00:09:18] Mark Williamson: So they, they serve that function very well. Um, we picked, they're, they're, we have found them to be incredibly nimble. 'cause we said, we said this is what we're building. So we're building what we call the Costco Velocity ad tech stack. We have many partners. Anyone that we bring in to integrate into our stack, we need them working off of our first party data and we need them working and being interoperable with all of our partner set. 

[00:09:41] Mark Williamson: So that could be our identity layer, it could be our privacy and consent layer, it could be our audience building layer, whatever it is. And so, A, any of our partners have to be agile enough and willing to play the role that we've asked 'em to play. And so symbiosis has checked all those boxes and has been great partners in doing that. 

[00:09:58] Mark Williamson: Um, their platform itself, [00:10:00] and I think the original idea from them was like, what does, what does Google PLA or search ads look like in retail media? Because when, when I first started trying to do this at the other place, 10 or 10 plus years ago, it gets really weird because you have an active digital marketing team that is buying keywords, driving kind of the core self-funded PLA strategy, because SEM is a big deal. 

[00:10:24] Mark Williamson: And affiliate is a big deal. And so what ends up happening is you're, rather than partnering in the ecosystem, you're competing and you're driving up CPCs and, and, and not able to coordinate. And so I think the, the original premise of symbiosis was to be symbiotic in that relationship. Where this idea of collaborative bidding where one plus one equals two and a half, where you take a retailer that is spending on certain keywords and a brand that's spending on certain keywords connected into a retailer's product catalog to, to be more competitive in the space and to avoid an open market [00:11:00] competition, which doesn't make, doesn't make sense for anybody at all. 

[00:11:03] Mark Williamson: And so. Um, we definitely see that it's a little different with Costco. Like we have no history of doing SEM. We don't do affiliate, we don't do paid media. And so this is a whole new ground for us. And so like what the thing that we're incredibly excited about is for the first time, brands can buy against Costco items in the Costco catalog and drive conversion at Costco, which is going to drive, uh, up, up funnel exposure, uh, side traffic. 

[00:11:30] Mark Williamson: Uh, probably warehouse traffic and then ultimately sales conversion. And so, like, we see a, a ton of head headroom here just because it's been, it's really been untapped, uh, for us as a retailer. And so the collaboration right now is using our data signals along with brand, uh, working media investments to drive demand. 

[00:11:48] Mark Williamson: Uh, whether or not we get into the collaborative bidding, uh, we'll we'll see. But we, but one reason, uh, we like symbiosis is we, we know we have the option to do that. 

[00:11:59] Todd: [00:12:00] Mark, one of the things that we go back, you know, a, a year ago today, I'm curious about and move into the Google PLA is an interesting one where a lot of retail media has been, how do I go up funnel more of funnel tactics, features, and you know, PL. Google Shopping, Google, PLA ads are, are seen as more middle bottom funnel. 

[00:12:21] Todd: What's your perspective on the landscape, you know, last year for this, this year, are you still seeing as much of an emphasis on trying to go more top of funnel, less, you know, direct response in what you're seeing in terms of the retail media landscape? And what's your perspective on, you know, the, the, the, the feeling in the space of should we be more top of funnel versus bottom of funnel as a, as a segment. 

[00:12:44] Mark Williamson: Uh, yeah, I think, I think the, the whole full funnel thing is, is evolving for, um. I think a, a big catalyst for this idea of even introducing upper funnel into retail media was because retailers wanted more and more ad revenue [00:13:00] and, and so like, okay, we gotta get to the bigger budgets. Well, the bigger budgets aren't buying down funnel or buying up funnel. 

[00:13:05] Mark Williamson: And so, uh, but it was always kind of an either or, like you're either selling yourself as a publisher and your lower funnel, or you're working with brands and agencies and you're trying to get up funnel. And, and that's fine. Like I still think that's an element of what's going on here, where you've got two different buckets with two different objectives, uh, spending in isolation on certain things, and you hope they kind of connect that you hope they meet in the middle. 

[00:13:28] Mark Williamson: What has really changed over the last year is this idea of a, of the connected funnel and using retail media, whether it's retail media data or retail media ad solutions for brands to connect their upper funnel. Campaigns with their lower funnel campaigns and the, the example was given to me, so I had a, a really enlightening conversation with the CEO, who helped me understand why this was important and illustrated how, why it's so hard and why few retailers can actually pull this off, even if they have technically upper funnel solutions and lower [00:14:00] funnel solutions. 

[00:14:01] Mark Williamson: The ideas is that brands are gonna do what they're gonna do. They have a story to tell, they have an upper funnel strategy. They're not gonna change that just for a retailer. But by part, by partnering with them on their upper funnel activities, it's not just about targeting the right members, but uh, but also understanding who has been exposed and then, and then taking what starts as a upper funnel, large, probably. 

[00:14:24] Mark Williamson: You, you know, uh, like a larger audience, understanding who's been exposed and then nurturing that same cohort down the funnel as they start, as they continue to make different, um, interactions with, with a retailer brand or with a brand. And the example that was given to me was if somebody sees a CTV ad for a particular item, maybe the goal is, is we want a frequency of of, of, we want 10 brand impressions with that person, uh, with our, with our video asset, either connected TV or online video or whatever it may be. 

[00:14:56] Mark Williamson: And then that cohort is now, has now been [00:15:00] nurtured. So if that person shows up on my website, that brand now may, it'll impact. Their strategy of how to use our publishing solutions because we can say that person we know, uh, we know saw your ad is now on my website. Do you wanna change? Maybe it's your bidding strategy or maybe it's your frequency strategy. 

[00:15:22] Mark Williamson: Uh, or is there a sequential messaging. Pattern that you want to hit. Where it starts as upper funnel, no co-branding to middle funnel. It's still a, now it's still a upper funnel asset, but it's now it's co-branded with a retailer. And then finally on a retailer website. It's, it's, it's heavily, you know, it's obviously heavily, uh, geared toward that, uh, that retailer, but it is conversion based. 

[00:15:45] Mark Williamson: It is buy now. It's not learn more. And so that to me is what is really changing, uh, you know, within this connected. Funnel, full funnel strategy and this like, it's very hard to do 'cause the signals have to connect all across the board, including channels that we [00:16:00] control or we're the publisher and channels that we don't control. 

[00:16:03] Mark Williamson: Which is why like we, we have put an emphasis on making sure that every channel we support is connected into our platform so we don't have signal loss. And so we can actually enable some of these capabilities that brands are asking for. 

[00:16:15] Tom: So, yeah, keeping on the, the brand interest and that's really exciting stuff. What you were just talking about. Um, I was just on a panel with Jason O'Toole from Gilden Haynes, and he was talking about persona threading, which sounded very similar to what you were talking about there, which is, you know, how do we get to sequential, how do we get to sort of actually using the data that's coming through all these touch points. 

[00:16:34] Tom: Um, super interesting. I also heard from some other brands about, you know, their reaction to Costco offering this new capability, and it was kind of like, does this remove the worries that I had about, you know, how do I figure out if Costco is performing for me? So regional tests on the Costco specific SKUs. 

[00:16:54] Tom: You know, using Instacart data. Does this sort of, you know, are you, what are you hearing? I don't know, it's brand new, but what [00:17:00] are you hearing from the brands about how this changes the relationship between you and the brands? 

[00:17:05] Mark Williamson: Um, not, I don't, I don't think a lot in terms of changing. I think there's a lot of general interest in what this can do to drive growth. And, and I, if I go into like the psychology of a supplier is, uh, you get into Costco and then you demonstrate the ability to drive sales velocity. And then that usually turns into good thing. 

[00:17:23] Mark Williamson: So either you go from a regional item or test item into a national item, and then maybe you become an everyday item, uh, or at least on some regular rotation. So that's really the, that's the psychology that we're trying to ladder into is if you, if you're lucky enough to get your item into Costco, what can you turn on in support of that item to get the sales velocity up to whatever the benchmark is, $500 per warehouse per week. 

[00:17:46] Mark Williamson: 5,000, what, whatever, whatever that that metric of success is that they've agreed to with the merchant that's going to turn on good things, that that's really what we're going for. And so like, it's not necessarily like this, this doesn't, I don't think change, [00:18:00] change the approach, uh, to how suppliers are trying to work with Costco, but it gives them more tools in their, uh, in their tool bag on how to. 

[00:18:09] Mark Williamson: Uh, increase the probability that they will have success because like the, the model for Costco is people come to the warehouse. They walk around, they find this great new item, they try it because it's at Costco and then they move on. Uh, and, and I think like if we we're really trying to say is like, that's, that's all good. 

[00:18:29] Mark Williamson: It's the right item. A buyer has made a bet on it. They've taken precious floor space to put this item in the warehouse and they hope that it sells. Can we help spark that, that. That demand, can we spark the awareness? And that's where, whether it's upper funnel or middle, middle funnel, can we pre pre-create demand? 

[00:18:48] Mark Williamson: So it's not only. The, the warehouse discovery where you accidentally bump into an item you never knew existed. Can we seed the market a little bit? And, and it's like, yes, we're talking about Google PLA today, [00:19:00] but like there's other ways of discovery. Increasingly, the LLMs are becoming part of that process, and we see that as an extension of this exact same concept. 

[00:19:09] Mark Williamson: Social media is the same type of thing, and so we see that that as an extension of this as well. And so like our, our, like the conversations we have with suppliers is how can you gin up demand in addition to the already massive demand flywheel that Costco creates just by existing and putting items in the warehouse because that works really, really well. 

[00:19:30] Mark Williamson: But can we grow a point or five points on the margin? Can we get the sales per warehouse per week up to that benchmark a week faster or two weeks faster by preceding the market a bit with some of this, this upper funnel demand? 

[00:19:43] Tom: So now we know why you call it Costco Velocity. 

[00:19:46] Mark Williamson: There's, uh, many, many iterations of the naming convention, but we are trying, trying, like this idea of sales velocity. Is it, it's Costco, it's the speed of sales that really makes this flywheel work. It's [00:20:00] that efficiency that's baked into the business. We're trying to take that exact same concept into the advertising space. 

[00:20:07] Todd: Mark. One interesting takeaway conversation and even naming it, Costco Velocity is. It's clearly a offering intended for endemic right endemic suppliers, vendors to Costco, whereas a lot of retail has been about whether it's top of funnel on endemic and to an extent, one of the dynamics of retail media versus traditional publishing is more limited inventory like we hear about retail media getting sold out, um, especially in store is a great example. 

[00:20:38] Todd: In-store retail media. We hear lots of stories of how that sells out very quickly and. The, so I think it does that lean into the focus of possible and retail media being more em focused than what others have thought about for retail media. And you know, do you think that if that's the case, is that, you know, a purposeful focus and, and where you'll be versus some of the [00:21:00] other, um, RMNs out there? 

[00:21:04] Mark Williamson: Um, just to repeat back 'cause you're cutting out a little bit there. Um, uh, just the premise of the question is, uh, is the idea of focusing on endemic about, about, uh, about working within the scarcity that we have in terms of available placements. 

[00:21:22] Todd: That's part of it. And, and does that endemic focus become something that is sort of more core to Costco as a result versus other arm ends, which might be a little bit more non-endemic focus in, in comparison. 

[00:21:38] Mark Williamson: Um, yeah, like I, I think our, our, our, uh, emphasis on endemic uh, advertisers is, uh, is the strict adherence to the. Belief that people join Costco and then they renew their membership because of the items that we sell. And, and the, [00:22:00] the more disciplined we are to that truth, the more likely we are to have retail media be a driver, not just a product sales, but of of membership as well. 

[00:22:10] Mark Williamson: And, and so we, we, we don't wanna deviate too far from the stories of the items that we sell. And, uh, and, and there's a, you know, there's several other reasons why non-endemic is, is somewhat prob problematic for us. 'cause there's not very many verticals that we don't compete in. 

[00:22:25] Tom: Yeah, I was gonna say, you guys have the travel and the hearing aids, so who else is left? 

[00:22:29] Mark Williamson: there's there, we have a very, very small list of non endemics that we, uh, that we, uh, would like to be our first partners in this space. 

[00:22:37] Mark Williamson: Uh, but, uh, but in reality though, we, we compete in, in so many different areas, uh, of the market that it becomes hard to identify a non-compete, non-endemic, which, uh, which is a, a bit of an issue, but, uh, but it's, it's something we haven't necessarily given up on, on trying to find what that right, that right mode is. 

[00:22:55] Mark Williamson: Um, but it all comes down to demonstrating member value. And non-endemic [00:23:00] just makes that very hard. And so far it has not. It is not elevated high enough to, to cross our hurdle to say, yeah, this is worth it. Even though there's definitely a lot of interest in partnering with Costco, whether it's, uh, within our own supply or, uh, or with audiences or whatever it may be. 

[00:23:16] Mark Williamson: Uh, we, uh, we have not yet figured out the right member centric reason for why we would do non-endemic. 

[00:23:22] Tom: So I wanted to go back to some of the shopper experience stuff that you were talking about before. You were talking about seeding the market. I think that's interesting. Um, but just for, for Costco shoppers. Um, you guys don't really have like a persistent cart. It's more like, maybe I have a list, maybe I don't, maybe I'm just going on a trip, you know, and I have some things in the back of my mind. 

[00:23:43] Tom: Um, like when you have a, uh, that single skew upstream one, I guess, are you looking for e-commerce growth right away, or is it more just sort of getting people to have better lists when they get in there? 

[00:23:57] Mark Williamson: Uh, I think that it, [00:24:00] this we, the answer may de depend on who the, uh, uh, who the stakeholder is. As a, as an advertiser, this is positioned as a performance product. We, we call it performance shopping ads. And so, uh, if, and, and it's, it's a, it's a CPC basis. And so the, the idea is, is that this drives sales velocity quickly. 

[00:24:18] Mark Williamson: That because this is lower funnel you're speaking to in market, high intent audiences that should drive conversion. Uh, the way we've set up this channel is to be able to, uh, attribute to both an online and an and a warehouse sale. Like we're, we're indifferent to where that sale actually happens. But our belief is that this is going to drive e-commerce traffic and sales primarily. 

[00:24:40] Mark Williamson: Uh, we may, we may surprise ourselves, we may find out that people that are researching mattresses or refrigerators or televisions are doing their research on, on a search engine and then ultimately buying in the warehouse. But, uh, time will tell what the, what actually happens there. 

[00:24:56] Tom: Well, I mean the, yeah, the velocity thing I think is a great name [00:25:00] because there's this worry initially where it's kind of like, oh, if I find the SKU online and it's at Costco, I can get it delivered, then I don't go on a trip and isn't, you know. But I've also heard, and I think when we were, uh, at NRFI heard this as well, is like, I didn't really get the idea that retailers sometimes don't like ads because it slows people down in the store. 

[00:25:20] Tom: And so velocity overall is, is a much bigger concept for a retailer than it is for media. Media people like us. 

[00:25:27] Mark Williamson: Y Yeah. And, and I think the, the idea of like this single item, like, like this, something I learned when I got here is that Costco is item merchants. It's part of that skew discipline of, of like, they're not category managers. They have a very select budgeted number of SKUs that they manage, which allows them to go incredibly deep on every component of that item, price and quality and, and, and, uh, and service levels and all the, all the different elements of that. 

[00:25:53] Mark Williamson: And so like, I don't think it's necessary. I, I think, I actually think it's on brand to be item centric [00:26:00] in the things that we're doing. And so we just have to believe that an item can drive a trip, which then expands into more items, of course. And so the item drives the shopping occasion. The shopping occasion drives the basket. 

[00:26:14] Mark Williamson: And so if I look at our e-comm business. For, for kind of the big and bulky area, there are a lot of one and done types of transactions. Like I am buying a fridge. You buy a fridge, I'm buying a mattress, I buy a mattress. So the e-comm shopping behavior is a little bit different, particularly in non-food, non consumables. 

[00:26:30] Mark Williamson: And most of the campaigns we think we're gonna be running through this channel are going to be on items. And so the, the thing for us though is, is that if, if all it takes is one set of tires or one mattress to justify the cost of membership. And so if we are helping members, uh, if we're reinforcing the value of membership by helping a member discover and save on an item that they need, very much, that's a big ticket item. 

[00:26:56] Mark Williamson: The, the likelihood that they are going to renew their membership is very high. And if they renew [00:27:00] their membership, then they're shopping. And then you get the, the impulse buy, you get the warehouse shop where the basket's fuller and fuller. And so like, like we have to, like, even if this is an item. 

[00:27:10] Mark Williamson: Marketing channel and it ends up being a item transaction. That's okay because that feeds the beast. That drives the rest of the behavior, which is, which I think is about bigger and bigger baskets. More and more trips over time. 

[00:27:28] Todd: One of the interesting comments, mark, is listening to Costco and still maintaining its discipline around what it thinks about the purpose of retail media is, and not chasing the shiny objects and even using something like PLA. When we originally talked, you talked about the challenges of of selling in different RN concepts. Costco and, and the leadership there. Did you have, what was their reaction to selling in Google or was there any concern, Google as this sort, again, walled garden, black box, sort of beast. And [00:28:00] as a follow up to that, as you think about other, uh, initiatives across landscape of, of RMNs, what are some of the areas that you're interested in pursuing or researching going forward? 

[00:28:13] Mark Williamson: Yeah, the, the, the selling process is always challenging and it depends on who we're talking to. So legal has a reaction and it has a reaction, and, uh, merchandising has an a, a reaction. Uh, this is actually a channel we were pretty excited for because. Uh, merchants fundamentally know that SEM is a thing that's happening. 

[00:28:31] Mark Williamson: So if you search for a refrigerator on Google today, you're gonna get a lot of products from a lot of our competitors, and you're not gonna see any Costco there. And so that means like we are absent at the point of decision and possibly the point of purchase, and we never even had a shot. The problem is, is that as a, as a, an organization that does not invest in marketing, like we're just, there's just this gap. 

[00:28:53] Mark Williamson: And so I think our merchants fundamentally know that this is a way of driving traffic and sales. So they're actually excited [00:29:00] about the opportunity. Um, you know, there's always that apprehension around like, well, what else could we be spending money on? And is this money gonna come back to us in cost? And all that kinda stuff. 

[00:29:07] Mark Williamson: So we have to have kind of our safeguards in place for how we make sure that these are incremental investments made by performance marketers. Who this is something that they regularly buy rather than having this be funded by trade. And so like, like, but, but in general though, the way that we have pitched this, like this is an offsite channel where the bulk majority of the investment goes toward working media so we can reach as many people as possible and sell as much stuff as possible. 

[00:29:32] Mark Williamson: Like this is not a, a retail media margin maker. And so we've really just focused this channel on, on driving, uh, awareness, traffic and sales. And, and, and I think like, and that's true of, of every, everything that we're pursuing is that, uh, the, the benchmark of success is, is this driving relevance for members? 

[00:29:52] Mark Williamson: Is it driving, is it driving sales, conversion, uh, something more than just ad revenue. And, and, and what that [00:30:00] ends up doing is it makes it, I think, more complicated to get some of these solutions out the door because if we can't. Target it and we can't measure it, then we can't go back to the organization and say, Hey, we did something good. 

[00:30:13] Mark Williamson: We did a thing that you're interested in. And so, um, uh, so like we, we, we continue to, I think. Make inroads within the organization because that centricity on member relevance and top line sales is something that we can all agree on. Um, and you know how we do it, how fast we do it, what the investments are required, what the resources are required. 

[00:30:33] Mark Williamson: That's all up for debate. But increasingly though, we feel like the tide, it has turned a little bit, the conversations are shifting more into, I see the value of this as a way of growing my business. Uh, and as, as we look forward though, I think we're, we're gonna continue to move upstream with this. Uh, we're innovating on our, our existing offsite um, solutions. 

[00:30:56] Mark Williamson: Uh, we're trying to crack a few more walled gardens, which again, is very hard. [00:31:00] Uh, 'cause you have to have exposure files, you have to have certain data terms. And there's just like the, these are big rocks that we're working to move. Um, but, but we, we, we remain committed to making sure that. The addressable aspects of the Costco velocity stack can reach as many of our members as possible, uh, on the open internet and in, uh, in, uh, in walled gardens as well. 

[00:31:23] Mark Williamson: So, a, a lot more to happen within our offsite pillar. Uh, we have an awful lot of innovation coming in our onsite pillar, uh, for the rest of, uh, the spring into the summer, uh, leading into holiday. So, uh, we. Uh, we're innovating a across the, across the stack and making sure that we are interconnected between the offsite signals and the onsite signals. 

[00:31:43] Mark Williamson: So, uh, a lot more to come for sure. You just have to say, subscribe to the newsletter 'cause that's where we'll make news first. 

[00:31:49] Tom: Yeah, no, and I think you're, you're setting yourself up to have a third and fourth, uh, appearance on the pod, so I love it. Um, wanted to close with a new type of question. I [00:32:00] shared this out on LinkedIn, so maybe you saw it, but. Um, what's the boldest either retail media or trade experiment in your vast experience in being in retail? 

[00:32:10] Tom: Have you ever run and to, you know, give you time to think, because this was at the end of the questionnaire. Um, examples would be like Sam Walton and Made in America, talks about how he brought in Zuri sandals for 19 cents Cool Price. Threw 'em on a table and they went like hotcakes. Dollar General started with pink corduroys, sold 'em for eight bucks, and then the whole town was like an ad. 

[00:32:31] Tom: Um, five below sends deflated basketballs when they ship because they wanna be able to ship more. So like cool stuff like that. Do you have something in your career that is similar where you had a lot of fun testing something out? 

[00:32:45] Mark Williamson: Um, yeah, so since I saw that question, I've been trying to, I've been trying to come up with something that would be interesting. Uh, and, and I don't know, maybe I'm not as bold as, as other, as other people or, or they just, they just seem, [00:33:00] uh, they just seem a little bit too, uh. Uh, I don't know. Too boring, but, uh, like, and this is completely honest, like there's probably a little too self-serving and a little too on the nose, but, uh, but like, I think this idea of bringing the best of retail media to Costco I is like right now, I guess the, the, the, the boldest thing happening from me and my team and, and this decision to meet the demand for transparency. 

[00:33:28] Mark Williamson: W by being transparent, not just about like data and making data available and dashboards or whatever. And, and, and performance transparency, which I think generally is what's being asked for, but also like, like foundational transparency, like how the sausage is actually made and, um, like which, 

[00:33:48] Tom: That is bold. No, no other, no other retailers like talking to that level of detail either because. They're not sure they can or they don't think they're allowed to. Right. I mean, 

[00:33:57] Mark Williamson: When, when I think that like Costco has [00:34:00] a code of ethics and one of them is respect for, uh, respect for our suppliers. And like, uh, like as I've kind of been ingratiating myself within the culture and they like, what does that look like in the retail media space? Because the relationship is entirely flipped. 

[00:34:14] Mark Williamson: We're on the selling side, not the buying side, but at the same time we're still representing the retailer. So there's still kind of this weird dynamic. In the relationship and so kind of like, like what does respect for the supplier look like? When I am actually the. Supplier, not the buyer. And like, and this is really what came to the forefront, was let's just be wildly transparent about what we're doing, what we can do, what we can't do. 

[00:34:41] Mark Williamson: And, and this idea of, of not waiting for perfection before we go out. And this makes my team a little bit nervous as we, as we're going through this, because we're not at Fulbright on every channel. Like even even with what we're doing here with Google, like we could have waited and we could launch. The symbiosis platform within Costco [00:35:00] velocity until we have Google and Bing and the CER and the social channels and everything else. 

[00:35:04] Mark Williamson: Like, like I said, no, let's get, let's get an MVP out there that is as connected as it needs to be, is as good as it needs to be. And let's go build on that and let's invite people in to be partners with us. And what'll happen is they'll get their fingers dirty, uh, as they're working with us. They're gonna have opinions. 

[00:35:18] Mark Williamson: They'll tell us where our, all of our, our scars are at. And it may, it may be great. It may suck, but we'll, we'll deal with it. And so I think that's the, that's the, uh, the objective of what we're trying to do is to show respect by being, by being blatantly transparent about what we're working on and why. 

[00:35:38] Mark Williamson: And, and it won't be for everybody. Uh uh, and we'll, we'll see hopefully more, more often than not, but like the, uh, the Costco ethos is bold by nature. I was trying to think of a Costco merchandising experiment that I could bring forward. The whole company is an experiment, like, 

[00:35:56] Tom: Right. 

[00:35:57] Mark Williamson: unless it's like paper towels, [00:36:00] almost every item is an experiment. 

[00:36:02] Mark Williamson: And if it, if it works, we do more of it. If it doesn't, we get out of it really quickly. And so, 

[00:36:07] Tom: Uh, you did experiment with the name premium towel, which is kind of a joke around my house, but. 

[00:36:12] Mark Williamson: I, no, no comments on that. But, but I think like the, the idea is like, like, like at the very, very beginning of this experiment, what I said was, we're gonna do a retail media program that reflects the Costco brand. And so this is just what came out of it. Um, I don't know if it was the design or not, but. Um, the whole thing is an experiment and if it works we'll keep doing more of it. 

[00:36:35] Mark Williamson: If it doesn't work, we'll stop. And this whole, you know, the PLA thing may not work. We will pivot or may not be good enough, and so we'll pivot. And that's, uh, I mean that, that's, that's about as bold as I can get. I'm fairly risk averse. 

[00:36:47] Tom: we're here for it. This is great. 

[00:36:49] Mark Williamson: Yep. 

[00:36:50] Tom: Todd, any any parting questions? 

[00:36:53] Todd: No, I'm good. It's, it's always the story of, of Costco and I do appreciate the [00:37:00] organizational discipline around the RMN and what they allow you to do and not do. Is very much in the Costco way. So it's a very interesting story where a lot of times RMNs are about chasing margin and dollars and you start seeing crazy things happen, um, you know, interesting in their own right. 

[00:37:16] Todd: But it's a very different model at, at Costco and it's, it's been interesting to, to watch the journey and it's, uh, growth. 

[00:37:24] Mark Williamson: Well, I, I, I'm, 

[00:37:25] Tom: yeah, go ahead. 

[00:37:26] Mark Williamson: no, I, I'm, I'm glad for the feedback and, and it is, it is a wild journey and I, uh, I was, had a one-on-one with my boss a few weeks ago, and I think I was complaining about something like lack of resources or, you know, just kind of like, like just challenges we were having, getting a few projects across the finish line and, and like, and part of me is just like, wouldn't it be easier if we just. 

[00:37:46] Mark Williamson: Get away with all these constraints. Like, 'cause we have the power to, we're not talking about technical limitations or, or regulatory limitations. We're talking about our own policies that are holding us back. And then as I'm talking out loud and he's, you know, either giving me a blank [00:38:00] stare or I could see his brow star and a furrow, I go, but because we're forced to do it this way, it re, it means that we have to be really sharp. 

[00:38:07] Mark Williamson: Yeah, we have to be incredibly creative, which I like some days is really irritating to me, but other days I'm still grateful for it because if not, I think we would take the path of least resistance and I don't know if we would like where that leads. And so it, it requires us to be as disciplined as the rest of the organization, but also I think it's, it's unlocking creativity that otherwise would not have happened. 

[00:38:32] Tom: Well, thank you for bringing the Middlemen along on that journey. Very exciting stuff, really like we 

[00:38:38] Mark Williamson: Thanks for having me. 

[00:38:39] Tom: All right, we'll see you next time. 

[00:38:41] Mark Williamson: All righty. Take care.