OLLY Agentic Commerce Optimization: MarTech & Digital Compliance Director Jennifer Peters On Building an AI Foundation the Right Way

Speaker A: I mean the opportunity to create all of these digital attributes now that you know, are beyond what Google has asked for, required or could read for all these years is really exciting too. And that's been, I mean probably one of the lowest resource projects we've worked on relating to AEO and um, probably in my opinion probably the highest impact.

Speaker B: Welcome to the Retail Gender podcast where we explore the intersection of retail, e Commerce and A.I. uh, agents, which we call retailgenic. Our mission is simple. We are shining a bright light on the new agentic AI uh shopping megatrend which will disrupt the trillion dollar digital retail market. The stakes are high. Google, OpenAI, Microsoft, Perplexity, Amazon, Apple, Meta, AH and Anthropic are all fighting to be the consumer AI shopping agent winner. There are going to be some tough decisions for online brands and retailers to make. What happens to checkout? How should payments work? Could you continue to invest in retail media networks? What does the world look like when over half of your traffic is not from humans but from AI agents? These existential questions are the tip of the iceberg. On this podcast we're talking to industry experts, analysts and influencers to get answers to these questions, identify trends and best practices. This is Retail Gentic, your podcast partner in navigating our AI shopping agent future. And now here's your host, Scott Wingo.

Speaker C: Welcome to this episode of Retail Gentic. Uh, this is, I know I say this a lot but this one's really interesting. Um, uh, we invite a lot of people to come on the pod that can't because they work in big companies and they're very um, the big company is very concerned about um, you know, talking about agentic commerce, revealing their strategy. They view what they're doing to be very strategic. So they're very, they want to be publicly quiet about it. Um, but one of the most, when we, when we talk to listeners and viewers like you, you want practitioners. So it's been very hard to get practitioners on, on the show. So I was reading an article about um, agentic commerce and um, this uh, this one executive, uh, his name came up a lot and she was sharing the best practices of what they do. So I took a shot and invited her on the show and she said yes. So I'm excited to welcome Jennifer Peters. She is the director of D2C. That's the part we're going to care about most. And then Martech and digital compliance at Ollie. Ollie is one of the, the um, you've uh, you've seen them before I'm sure. And It's a CPG company and uh, she is very gener the best practices that she has discovered in this role. Um, I learned a ton from this conversation and it's very validating to about in September of last year. So let's call it six months ago which in AI time is 10 years ago. But six months ago uh, I put out the framework of agentic commerce optimization of some of the things you should do. You've all seen the stair step diagram there where it's kind of like make sure your site's crawlable, make sure that you have done your basic attributes and beef them up, add some attributes, make sure your reviews are getting in to the engines and then make sure you, you have a frequently asked question and then beef up that frequently asked question. Um, and it's really cool to hear someone actually doing that in the real world and it really driving a lot of impact for, for, for the company. So um, there's a lot to unpack here, a lot of best practices. She's um, like many people that are from the Silicon Valley Bay area, Jennifer talks fast. So um, you may not want to speed this one up if you usually do. But uh, if you can process that information that fast, that's fine. Um, so I think you're going to really enjoy this. It's very, very strategic and practical, uh, and tactical at the exact same time. Here is my conversation with Jennifer, the director of DTC at Olli. I hope you enjoy it as much as I did. Welcome to this episode of Retail Genting. My guest today is Jennifer Peters. She is Director of DTC MarTech and Digital compliance at Olli. Jennifer, welcome to the show.

Speaker A: Thanks so much for having me. I'm so happy to be here.

Speaker C: We haven't had the pleasure of meeting before but I was reading an article about uh, all the cool things you're doing and I said I've got to get her on the podcast. And you were very, very kind to take my cold outreach and uh, agree to come on the pod. So I very much appreciate that.

Speaker A: I don't know if it's cold outreach when everybody knows who you are. It's, it's more just like an exciting day.

Speaker C: You know we weren't friends yet but now we are. So you never know some, you know how folks are going to react. Well, awesome. Before we, before we jump into the. And I know you, you're uh, I could just tell from the article you're, you can, you're a hands on type person which I love. Um, that's what's fun about E Commerce is we, we um, we're not super high level execs. You have to be hands on in our world and like understand all that stuff's wired together. Um, but before we get into the down into the, the metal, um, give us a little bit about your background. How did you, is this, how long have you been here and where, where did you work previously?

Speaker A: Sure. So I've been at Olli a little bit over five years. Um, I came on after the Unilever acquisition so I didn't, I did not live through that. I came on um, kind of mid Covid. Um, we've changed a lot even in five years. We're only 11 years old. So um, you know sometimes it feels like that, sometimes it doesn't. Um, but before that I was at an agency. Um, I'm a Shopify developer so I was working on lots of different stores building e comm experiences for different brands. Most of them in the CPG food space, like shelf stable food space. So it was a lot of fun. Got to see a lot of weird problems and strange products and new challenges. Um, but before that I um, did a couple other different things. I was a director at um, cokesberry which is like a church supply company in Nashville. Um, as a director of D2C um, e commerce and marketing there. So that was fun and was just telling you earlier like we're old timey uh, Channel Advisor customers. Even with a original contract that was done via fax. So.

Speaker C: Wow. I mean I probably signed that one if it was by fax. I, I'm sure I signed it.

Speaker A: Yeah, I wish I, I wish I had it still. I don't. But um, yeah, I've been in the business for a while, worked on a lot of different brands. I feel like this maybe is the most exciting time I've seen yet for D2C with, with everything agentic and everything AEO and everything changing week to week. Um, you know I think we go into this business because we don't want to be bored and we like to learn constantly and that's what we're doing.

Speaker C: Yeah. And did you pick up, you kind of threw casually in there that you're a Shopify developer. How did you pick up and sharpen your Shopify skills?

Speaker A: Well, I got tired of developers telling me no and, or that it would take six months to do something. So I mean I'm self taught. It was like before the days of Claude and uh, OpenAI. But um, am I the greatest Shopify developer in the Land, probably not. But I'm pretty functional and built a lot of stores. Um, and I enjoy it because it's, it's a lot of problem solving. A lot of, um. One of the things I do on the side, I do re. Platform a lot of brands off of legacy platforms onto Shopify and I really do enjoy that because I do feel like Shopify is the democratization of E commerce. Anybody can do it. Now if you work with an agency that makes your store so complicated that you can't do it, that, that's, that's something that personally bothers me. So what I, my, my kind of bread and butter is building stores that brands can manage themselves.

Speaker C: I'm going to use my predictive skills. I'm going to hazard a guess that Ollie's on Shopify.

Speaker A: We sure are. Yes, we are.

Speaker B: Whoa.

Speaker C: How about that? I'm pretty good at this. Toby Lucky would be proud. We're going to go on a little side quest here. And so we're in the little Shopify side quest. Have you, you know, have you now connected your Claude and your MCP over into Shopify or you're like leveling up and you're in kind of AI God mode with this thing at this point?

Speaker A: Yes, yes. And it's really almost terrifying in a way. Just, it's overwhelming.

Speaker C: Too much power.

Speaker A: Yeah, it is. It's too much power. And a lot of these things that in the past, you know, you've been limited by your cms. I think it's a reason a lot of brands were drawn to Headless experiences, um, because they wanted to have all these choices. But, you know, cost of ownership of Headless, pretty wild. Um, and not for everyone. Um, but now it is like anybody can do anything. It's like the ultimate democratization. Like you don't even have to be an engineer to build crazy stuff now. And it's, it's pretty cool. It's really exciting.

Speaker C: Are you totally Claude pilled and it's Claude all in for you and then, uh. Or are you also, you know, do you still kind of go over to the other side and look at Codex or what. What is your. What's in your. Um, uh, super AI Nerds would say what's your stack? So what, what kind of tech stack are you using here?

Speaker A: So I would say on my, on the freelance side of my life, definitely all in on Claude all in on everything. Um, we are a Unilever brand at Ollie. So we have a little more governance. Uh, a lot more governance. Um, and I think rules and requirements around what we can and cannot do, what we can and cannot integrate. Um, so we're moving slowly and thoughtfully into that space, uh, making good decisions, hopefully not, you know, not just a bunch of short term decisions, but really kind of trying to look at the full landscape, uh, where we are now, what we're using now, where we think we'll be in a year. Um, and honestly not over investing in any one thing at this point, just because it really does change week to week. And if you go all in and all in on something and then a week later it's like irrelevant. You know, you waste a lot of money and resources. So we're just going carefully, I think, very thoughtfully.

Speaker C: Yeah. So then within Claude, there's uh, just kind of using it in the chat mode and then there's cowork. Ah, and then there's code. Where, where are you on that spectrum of how you're, um, and maybe this is a question for the freelance side, but when, when you're interacting with Shopify and doing some of your hardcore stuff or which of those, which of those are you're working on?

Speaker A: It depends on the brand and it depends on the store. Um, one of the things I think about a lot as we talk about AI and different conversations is that there are a lot of brands that are

Speaker C: far,

Speaker A: um, more uh, traditional and rudimentary than you would ever believe. I mean there are brands, I've worked on a brand that is a pretty big brand who was like, we don't care what our store looks like, we just need the cart to work. And so I think, uh, sometimes the table stakes are so low for some brands that they're not even halfway there yet. And the problems you're trying to solve in those brands are very basic problems. They're not necessarily like, how do we take this integration to the next level? Like, how do we do all these crazy things? It's really more of just like a function of what the brand needs from you. So it really depends on the brand. I mean, I've got a couple that are very future facing and really excited and want to see, you know, like, how, how much have we increased our visibility by doing this, this, this and how do we know, how do we know that this is true? How do we know these measurements are true? Um, and then I've got, again, I've got brands that are like, can you just make sure our site works all the time? Um, and so there's quite a scale there. I think there's a wide spectrum in there. In There. Yeah.

Speaker C: All right, SideQuest is officially over. Um, I think most people know about Ollie, but give us like just for in case someone's like, uh, you know, we have a lot of audio listeners and they may not. So it's O l L Y. Um, and so tell folks, uh, how you describe the brand at a, like a cocktail party.

Speaker A: Absolutely. If you have ever walked into a Target and you have turned right and you have walked past the first end cap in Target, that is where we live, that is our homeland. Um, we're square bottles, gummy vitamins. Uh, we have a lot of different products but we're known for our gummy vitamins. They taste amazing, they work really well and we do have like distinctive packaging. So that's usually I think what people, um, when you say like we're in the square bottle, they're like, oh yeah. Oh, okay, got it. Um, there are a lot of other Ollies out there. There's dog food, there's like a toothpaste company now. Like. But um, we are O l l Y and um, we're, we're in the supplement space just having a great time.

Speaker C: Yeah. And then when we talked about your role, D2C, I understand that's, that's my people and then Martech. So um, you know, uh, is that meaning we're doing a little ad tech in here and you're, you're kind of uh, you know, into the retail media network world or what's embedded in that part of your title?

Speaker A: I try to stay out of paid. I do get involved in it sometimes. I think, uh, most of it is really DTC and uh, data related. I do a lot of first party data work, um, you know, kind of integrating that information and data into other pieces of the business. We are primarily a CPG brand at the end of the day. So the majority, the vast majority of our revenue comes from in store, comes from mass, Walmart, Target, grocery drug. Um, D2C is a pretty small part of our business overall. Um, but yet still important. Uh, Amazon also very important. Um, and now I think what's really exciting is that you know, before a year ago it's like, oh, DTC is just, you know, I don't know, 7, 8% of our business. It's not that much whatever. Now it's the point of entry for discoverability for the brand as we move into this world of LLM discoverability. And so now the stakes are much higher and there's a lot, a lot more, um, eyes on that part of the business. Just because you know, that is what, you know, that's what people. That's where people are going to go when they start asking questions, to chat GPT and like, how can they do this? What should they take? They can't sleep. What should they do?

Speaker C: Even if the consumer doesn't go there, whatever agent you're using. And if you think about, okay, I'm an agent and I'm trying to help someone find vitamins, and I need to find the authority on this thing called Ollie. Where am I going to go? I'm going to go to Ollie. Uh, and if I can't find what I'm looking for, then I'm going to go to, you know, Amazon and they're going to block me. Then I'm going to go to Walmart and Target, right? And, uh, um, as far as I know, I know Walmart's like throwing the doors wide open. Target's kind of like in the middle somewhere. Like, you know, they're, they're kind of like frenemies. Uh, and then, uh, yeah, so Walmart's on the forefront.

Speaker A: I mean, it's. Honestly, it's really impressive.

Speaker C: Yeah, yeah, they've done a great job. But this is like the most important part. You got to it very quickly, which is awesome. Um, when I talk to cpgs, there's this. It takes two or three conversations and then finally they say, I've literally had them. Um, there's like this cognitive dissonance that hits and they're like, you mean this website I've ignored for so long and I threw a where to buy button on there and no one cares about it is now the most important thing? I was like, I don't know if it's the most important thing, but it is the most influential if you want to solve a lot of your nightmares. You're telling me about the crazy Reddit stuff and all this kind of stuff. You, as you are the brand authority via your website, and that is your, you know, on that PDP now is your mechanism for countering all that. And it. There's finally an awakening there. And then they're like, well, that's going to be, you know, uh, and then they realize that they actually have a fair amount of control over that. And then they get kind of excited. So, so a lot of them are kind of like, there's this hopeless, like, oh, my God, I'm gonna have to go to Walmart and convince them to change this thing, know Sparky, in some, some holistic way, and I'll never get them to do that and it's like good news. You, you, you know, Sparky is just an LLM and it's consuming the same information. Even Sparky may. Your. Your site actually matters to Sparky and Rufus, interestingly enough. And a lot of, a lot of people kind of. It's really hard because we've lived in these wholesale D2C lanes for so long. There's this thing that cuts across them, uh, and it is the data from your website, which pretty much pretty wild.

Speaker A: It really forces you to look at your business in an omnichannel way. Even if you didn't want to do that or if you're not internally structured to do that. A lot of brands are not where, you know, there it is very siloed by channel and the channels. There is some channel competition and channel friction. This, this almost removes that entirely from the equation. And I feel like brands have to react to that quickly and, and restructure in a way that supports that because otherwise you do kind of have this like, infighting and it's pointless because it is definitely an all boats rise kind of world right now.

Speaker C: So you've obviously ended up at a great end point. But let's talk about your agentic commerce journey. How did you, how did you get here? Did someone, uh, at Ollie say, hey Jennifer, our Google traffic's a little weird. Can you go look at this? Or did you, uh, you read retail gentic and you're like, whoa, I need to hop on what Scott's saying. Or like, what, uh, uh, leading the witness. Or, um, you know, how did you, how did you realize, uh, that this was something that as the DTC practitioner, you needed to get in front of?

Speaker A: I mean, I'm out in the world a lot. I'm at a lot of conferences. You know, I feel like you're constantly in the space consuming information like, like podcasts and in webinars and uh, email newsletters and all of these things. And it's hard not. It's hard to miss that. Like, it's hard to miss like, whoa, this is the way we're going. And I think it really came to head probably like about a year ago, like mid, maybe mid summer last year, where it was kind of like, okay, now it's not just talk anymore. We've got to figure out an actual plan with tactics. And um, you know, that's when you start looking at like, what are the controllables we can control. Like you said earlier, it's. It's your site, it's your content, it's content that lives on your site. Um, there are things that are harder to control, like crazy Reddit, um, you know, like some of the things that exist out in the world that people are saying about your brand that's less controllable. Um, but we, you know, we made it a priority to really focus on the things that we own, the spaces we own, the content we own. And there's so much of AEO that is really just doing business well. It's really just good SEO. So we started with that because we didn't want to go overboard in one direction, in one space and then find out that that was going to be irrelevant in the future because it really just felt like, I don't know, nobody really knows what we should do next. So let's focus on the things we know we can do that will have an impact for LLMs and for humans and make the site a better experience. Clean up our schemas, clean up our data structure, um, you know, all of those things that are just going to be good for your brand anyway. So that's where we started, um, really, really in that space, uh, restructuring some things like putting um, putting content into FAQs I think is such like an easy slam dunk that any brand can do. And one of the things we've done is we've, we've added FAQs to our PDPs. And honestly it's the same information that's already on the page. It's not even new, it's just in a different format.

Speaker C: It's in a bite sized chunk. Yeah. And it's uh, and it's packaged in a nice little like, you know, boom. You have this question, you know, are we vegan? Boom.

Speaker A: Exactly. I mean you could have read the paragraph and found that out, but not everybody does. And I mean LLMs obviously love FAQ format, but what we found is that humans like it too. So you know, we're kind of measuring things on both sides of like what's impacting LLM discoverability, but then also what's impacting human understanding and education around like what our products are because they're supplements. You have. Not everybody knows what everything does. So there is definitely an educational component to what we sell that's really important. So measuring both, I think when it, when content is expensive and resource heavy to create, the best place to start is in the space where it checks both boxes. And that's what we've really been focused on, I think like this first, you know, this first year into this new

Speaker C: AI world phase one Is, um, take your existing content and gussy it up, um, get it kind of like into the right format. Frequently asked question is a very popular one. Um, uh, you know, another best practice is take your attributes and make sure they're in metadata. Did you do that? Yeah. Okay. As a Shopify developer, I had a feeling you were all over that. And then, then they said, well, you can do that. But then there's this, and I forget the Shopify vernacular. There's a, there's an agentic product catalogy thing they want you to put it in. And I assume you've done that too.

Speaker A: It's actually really, really cool because that, that is my, that is one of my favorite things to talk about. But I'm gonna tell you what, it's the least exciting thing to talk about internally. Like, people are like, that is the most boring thing I've ever heard.

Speaker C: Well, your people are here. Let's talk. Let's talk.

Speaker A: Yeah, we worked with a partner, um, that kind of specializes in catalog feed enhancement. And that was I think like such a huge unlock for us because we could do that, but it would take a lot of time and we're highly regulated so we have to be really, really careful about what we say, um, claims wise and kind of all of this stuff in the science space. We just have to be really careful. Um, and basically this partner could ingest our public catalog and create all these additional attributes for us and then just literally push it back into Shopify. And none of them are really customer facing. And again, a lot of those things, they're things that already exist, but they're not things that Google ever cared about. Like, Google never read if you were, if your product was NSF certified or not. They wanted to know what color it was or like what size it is. So I mean, the opportunity to create all of these digital attributes now that you know, are beyond what Google has asked for, required or could read for all these years is really exciting too. Almost scary because sometimes it's like, oh my, oh my God. Like the attributes are endless. Like we could put anything in here. Um, but we actually work with a partner that did that for us. And so they kind of ran it through their, their model and made suggestions on like these are the attributes that we think will move the needle on discoverability for this product and this product type. Um, and that's been, I mean, probably one of the lowest resource projects we've worked on that relating to AEO and um, probably in my opinion, probably the highest impact.

Speaker C: Um, so there's also a million vendors that do AEO GEO and this kind of thing. Um, you know, the biggest one is Profound out there and they've got this, uh, but you're inside Unilever, so I imagine that may actually live. What people tend to realize is it actually lives up, up a level into like the bigger marketing. Um, and it's usually company wide, not brand specific. Um, did you guys get pulled into that part of things or does that happen up in the parent company?

Speaker A: Actually, we have our own instance of Profound via Unilever and it is pretty exciting. Um, it's, it's really interesting. I think it's, it's cool to look at because it is honestly, their UI is like beautiful. Um, harder to take that information and be like, okay, here's the checklist of things we're going to do.

Speaker C: Um, yeah, it's not super actionable.

Speaker A: It's not, it's hard. It's, it's. Somebody has to analyze and digest that and kind of regurgitate it into tactics and you have to have somebody that can do that for you. Otherwise you'll literally just look at it all day and be like, that's so cool. That's so interesting. Um, it's got these, you know, beautiful logos and everything. Um, and one of the things I do like about Profound, um, is that like our business, we're in supplements. So there's a lot of different segments. There's like Multis, that's one group of competitors. There's Gut Health, that's a completely separate group of competitors. Women's Wellness, like, every segment that we're in has its own sort of set of competitors that are very different. So it's cool because it does break that down by segment. So you can see kind of like, here are all the segments that we play in. Here's where we're winning, here's where we need work, here's where we're getting destroyed by our competition. Um, and here are some segments maybe that aren't priorities for us that we can kind of back burner for now and really focus on the ones that are like, win categories for us, Us. So I do like how it breaks that down. Um, and it is really interesting, but it's also, I don't, I don't know if you have, you've talked about this in the show before, but their data sources are very mysterious. And, um, they have hundreds of data sources and they're all under NDA. So it's like you're doing, you're using this, but you're like, but how do you know? Like how do you know?

Speaker C: Yeah, they use panel data and panel data is what we would, what the kids would call super sus. Um, everyone says it's like super permissioned but then as you dig into it, the, the, a lot of it leaks through, um, Chrome extensions that have one function and then they're grabbing your data or Android devices that are doing stuff. So yeah, that's where it gets a little fuzzy.

Speaker A: It is. And we have, we have, we work. I'm not going to say the vendor's name. We work with a vendor that does data we love. They're great. And we found out that they are one of Profound's data sources and we're like, oh cool. Okay. So I think there's a lot of trust but verify in this space. Um, which is why again, like, I think we've sort of dabbled in different areas because we want to validate these opinions, experiences, suggestions against each other. So like if we see and profound a specific recommendation, if we see in Content square a specific recommendation and we see it also maybe in an envy, like we know this is probably good and it's probably one that's good to take action on. Um, but again, it's like everybody's data sources are different and you really just kind of have to like look at the resources you have and figure out the best way to spend them on what you have to work with. So that, that's kind of, I would say, our. Been our approach. We're very thoughtful. We want to stay human centered. I think that's a really important thing for us. Our products are very personal, they are ingestible, they impact your health and your kids health. We don't want to play a game. Like we want to make sure that the product that the customer buys is the right product for them. At the end of the day that's, I mean that's kind of how we, I would say we ground ourselves in, in that human centeredness as our approach.

Speaker C: Yeah, a lot of the aeo, this is not a profound thing, but uh, some of them, some of them, you know, they, they kind of, they kind of have these best practices that most brands I know wouldn't, wouldn't do. The best practices, they're like kind of, well, go create your own Reddit account and put a bunch of good stuff out there. No. Or they'll say take your customer service people and go, go engage with the Redditors. And we all know that that does not work. It's like just Hitting the hornet's nest with a, with a flaming wine.

Speaker A: You cannot control what you put on Reddit. I mean, uh, there are brands, I feel like, that really can engage on Reddit in a productive and cool way. Like, I have a friend that works at Sonos, the speaker company, and they have this massive Reddit community because it's like, people who love their speakers, people who are, like, trying to do this installation thing, or maybe they bought some parts from somebody, but they're missing some pieces. So it's like a practical application on Reddit. It's not just like, I hate this or I love this. Like, it's really a full community. Not every brand can do that. Like, not every brand's product fits that mold, but they do. And that's great. But, you know, it's much harder for other brands that are just like, do you like this or not?

Speaker C: Sa.

Speaker A: We use Bizarre Voice, um, because we're so. Yeah, we do. And we're very, again, we're very, very Omnichannel. So Bazaar Voice, like, that syndication piece is really critical for our business to be able to send those reviews to Walmart and to Target and our other retailers. So, you know, Bazaar Voice is our choice for that, um, just because of the syndication piece. But yeah, I would say, like, they've been really good partners on kind of getting with their customers and making sure that they are doing it correctly. Like, they want to make sure that, like, we were getting the full value of the way their product works. So that is something I was really impressed when they reached out to us and they were like, hey, we want to make sure that you're utilizing this, this, you know, kind of structured yet unstructured content, um, for aeo, like, let us help you figure out and validate to make sure that you are using it correctly, which I think is really cool. Like that to me, seemed like, very proactive. Again, we have, we have some products that are 10 years old, and Women's Multi is one of them. So there's a quite a few reviews for it, and some of them are really old. And, you know, and LLMs love recency. So that I think is, you know, an advantage there because then you don't end up with someone who bought it 10 years ago and was mad about it then. But, um, I do think the recency, sort of like mental filter helps a lot there and kind of filtering out any products that maybe have just been on the market for a really long time.

Speaker C: Cool. Anything else on that before we go, uh, into. So this is Kind of like the first wave of things you did and anything else you did in kind of like that first wave.

Speaker A: Well, you know, one thing I just want to say I. It's sort of related, sort of related to our philosophical approach to AEO at Ollie, is that there are, and I've seen this in a lot of brands, like there's a lot of pressure to do something with AI and there's a lot of pressure to go back to your board and be like, this is what we're doing with AI and this is how many monies we made. And sometimes we're overlooking like basic integrations and automations that like we probably should have done like five years ago. So I, I don't want to overlook the tools that you have and making sure that those are integrated with each other in a way that makes sense because then if you try to add AI on top of that, like everything's gonna be disconnected. So, um, I think most brands have opportunity that like predates AI to integrate and to kind of build out these like workflows and automations and things that are completely unrelated. Like maybe they're your, you know, they're your creative workflow software and your dam and like maybe those aren't connected. You know, there's, I think there are a lot of things, especially when you're in a business where channels can be kind of siloed, where you have things that are just wildly disconnected and especially if you're omnichannel, I think it's worth it to take a good hard look everything you have in your like creative stack, your marketing stack, your paid stack, like everything and make sure those are as connected as they can be before you try to like sprinkle AI, uh, magic on top of it.

Speaker C: If you're clever, you can use AI as an excuse to get it prioritized.

Speaker A: That's true.

Speaker C: Within reason, yes.

Speaker A: That is absolutely true.

Speaker C: So you did all this and you put it on your PDP and then let's talk about measurability. How were you able to measure that? This was getting into the engines that you connected those dots.

Speaker A: We do use profound. So I think that's, you know, being able to measure kind of that real time impact, whether you believe the measurement or not is really helpful. Um, at least directionally it's going to be mostly correct. Um, and we use Invive as well. And so we're kind of using both of those to measure impact by product. So that is helpful. I, um, there's so many things that I'm skeptical about in this space that, you know, our feeling is like to not get distracted by short term results, short term impact, and to stay focused on kind of the long game of what we are ultimately trying to accomplish here. Um, so I, I think that, you know, it's easy to get reactive in here and build a deck that's like, oh man, look at this. We did this and this happened. We did this and this happened, we did this and this happened. But really we have to look at it from a much longer time span of like overall impact and looking at like brand awareness and household penetration and kind of these other, these other measurements we've always had, we had, we have OKRs that we set for the business at the beginning of the year. Those are still our uh, okrs. Like we're just using AI to get there faster and to get there better and more efficiently. We're not changing the goals because we have AI, we're keeping the goals and we're just doing a better job getting to them. And so I think there's like a constant philosophical check in that you have to have within a brand and within the group that's working on AI and in your brand and just make sure you're making decisions for the right reasons and not doing AI to do AI, which is like my biggest fear. Um, because then you make dumb decisions and then you're not really thinking long term, you're not really thinking in a customer focused way. And there are so many things you can do. But like at the end of the day, do those really matter to people? Because people are using LLMs. LLMs are going to know if you're delivering bad results or results that are irrelevant to people because they're not going to buy or they're not going to click. So all of that stuff is going to come back around and get you. If you are taking shortcuts, if you are kind of trying to game the system or play, play some sort of game. It is like the old world of SEO where people were putting like white type in, you know, on pdp.

Speaker C: So like keyword jamming and all, all the things.

Speaker A: Yeah, gross. Like that's what we do not want to do. Like I feel like that's the cautionary tale there. Like let's not do that, let's just do it. Well, make good products, make sure everything is clean, make sure your house is clean and you know, then do the things that you can do to get that stuff read well by LLMs.

Speaker C: M. Yeah. On the measurability, are you Thinking of, um, you know, so, so number one, you don't want to do all this work to improve your product catalog and your PDP is hindering it in some way. So there's like a, there's like a crawlability of your PDP and your basic AOGO tools. Very good at just kind of telling you that or not. And then, um, now are you actually putting active prompts in there and looking for the women's um, gummy and making sure that your AI visibility is high or you're just kind of like seeing it in the panel data? Um, like an AI visibility score. Get a little more specific on that.

Speaker A: Yeah, I mean I would say we're doing both. So you know, in the Shopify knowledge bank you can do a lot of that kind of like pre prompting type of activity, but in Profound, you can do that too. So, um, I think that is really important to make sure that those things are sort of like going the same direction. Uh, Profound is probably more detailed in how you can, like I said earlier, like how you can break it down by segment. Whereas Shopify is more kind of like here are all the, here are prompts, like here, here are some things, you know, and they're kind of all mixed together. Um, and, but I would say too like, you know, our customers, millennial women, like the vast majority, and the agentic traffic and agentic commerce we've seen so far is really small. So I don't, again, like, I don't want to go crazy doing one thing, but I also want to make sure we're building. Like we're very foundation focused right now. Like what is going to be the foundation of what Discoverability is going to look like in the next year? Like, are we doing, are we executing the basics well and consistently without getting a little too kind of bogged down in the details and you know, I mean building, building out like an AI agent that can manage your inventory and place orders for you. Like that's great and that's super convenient and wonderful and other things you can do with AI on your site outside of, uh, Discoverability, just to make things easier for yourself. Plugs great into Shopify. It's, you know, works, works great. But uh, again, like executing the basics I think is still the most important thing.

Speaker C: Yeah, so many people haven't that it actually is you, you just doing the basics well and tightly. It makes you a leader. So that's, that's. We're kind of in that early phase where all you have to do is that. And it also, um, it starts compounding because you get, you get into the, you get into the web search part now and then in six months they're training on that data. Well now you're in all the LMS long term memory with the. Right, all the, all the right. The payload of really good product information. Um, um. And it's like a, it's kind of like something in your short and long term memory is an analogy I use. And if it's in both then you'll kind of like remember it forever and then it starts to leak into the non D2C part. Uh, have you actually seen any evidence of that? Where um, and I'm sure your, your wholesale colleagues come to you. Another thing is you've probably established I'm going to kind of be predictive again and say inside the company everyone's like who knows anything about this stuff? And they're like well call Jennifer. So you're probably the, you're probably the go to gal on all things um, AI at this point and I imagine, you know, so let's say there's someone in Ollie that owns the Walmart and the Amazon relationships. We see this all the time where um, the artist previously known as Rufus. Now I'll just use Alexa. Will, will, will say some pretty wacky stuff and then Sparky is like kind of a bit unhinged on what it will say. So they'll, they'll, they'll be running around and they'll notice that and they'll probably come to you. Are you starting to see evidence of this having an impact outside of the DTC part and over on the wholesale side?

Speaker A: I think a little bit. Um, you know, I mean again I, I would say I feel like a lot of CPG brands are far more traditional than people would think. Um, and buyers at Walmart and buyers at Target are a lot more traditional than people think. And there are incredible technical, technical teams at both Walmart and Target kind of working on these projects but they're not always as connected to like the day to day business of buying as you would think. So I think it's, it's still developing. Um, I mean we're very close to both Walmart and Target so when they do well, we do well. So I mean I think we want to support even in discoverability, like if it's, if we're coming up on Walmart.com and chat GPT like that's still a win. We love it. That's great. Um, and it's also like kind of a, a touchy area to navigate, I think too, in your relationships with retailers of like, are you supporting their business? Are you supporting your D2C business? Are you supporting your Amazon business? Because I think your retailers can get competitive in that space even if you are not, even if you as a brand or not. So I mean, I think it's like a, a careful space right now to navigate. Just make sure you're trying to kind of keep everybody happy and make them happy with the results that you're showing them, but not necessarily, necessarily kind of trying to like cross those lines.

Speaker C: One example that again, kind of within this realm you can control. I like your framework. There is. You've, you've done all this work on frequently asked questions. Take your facts and make sure they're getting into the PDP on Amazon, Walmart, Target and all the places you're selling. Have you, have you guys kind of done that step or is that maybe a phase two?

Speaker A: We have. No, I think that's, that's a great, I mean, I think content wise, making sure that you have, that you're, you're leveraging the same content across channels is really important. I mean, Amazon as well, like, if you've got something that's really driving discoverability in one channel, you want to have that in all channels. So that, I think that's another place where some of those silos have to break down a little bit, um, in terms of content and making sure again, like you're kind of all playing on the same team that you're all sharing what's working, um, across channels, um, and also testing across channels. Like our channels do not all have the same exact customer profile. So you know, there are going to be places where some things work really well and some things don't work really well. And there, there are channels where we do really well with certain segments. Like we sell a lot of Maltese, but we don't in this channel. So I think we kind of have to look at that too of like, what does the business overall like look like and are we supporting the things that are the strengths in that business by channel? Um, and that sort of takes some of the competition and I think and friction out of it too. When you're like, hey, look, this is the product line we do the best with at Walmart. Let's really focus on driving this on Walmart and on this channel. We're going to really focus on this segment because it's really strong in this, you know, in this channel. So some of that, I think you kind of have to like, you have to break down the business in each channel, too, and then figure out how you're going to support that by channel.

Speaker C: I've been working this for 30 years channel. Yeah, there's not, There's. There's kind of definitely cross channel optimization, but at the end of the day, if you really want to move the needle, like, really, really understanding the audience, it's all consumer behavior. And every consumer in every different environment is very different and a different profile and.

Speaker A: Yeah, totally agree.

Speaker C: So that's kind of where you guys are next. What's kind of on your, you know, Jennifer's plate? Uh, wish list for, for, you know, as we, you know, for the rest of the year. Like, what have you knocked off everything? Or what are some of the other big chunks you want to get to?

Speaker A: Content's slow. You know, it's hard. It's hard. Takes time to make content. So I think we have a lot of Runway still ahead of us on the things that we can control and the content we want to create. Um, I think one of the things I'm seeing right now that I find really, really fascinating are some of these brands that are coming out and saying, like, we don't use any AI in our creative. And that's kind of like becoming a selling point. And I don't know if you saw Ronny Chang's, um, commencement speech at Harvard, where it was just like, fai, like, that was like the whole theme of the conversation. And it was kind of like he was saying to them, like, you're like, the mandate of you genius people graduating from Harvard is to destroy AI. And so I think this, like, backlash and sort of this counterculture AI piece is really fascinating. Um, you know, again, I feel like we're in a really safe space because. Because we're, like, doing things that make sense, but we're not, like, going too far. Um, and I think there are some brands maybe that have gone a little bit too far, especially on the creative side. So I am kind of like, waiting to see what that backlash ends up looking like and if it gains traction or if, like, the convenience sort of takes over, um, from the, you know, one of the things he said is, like, the journey is not just how we acquire skills. Like, the journey is the purpose. And so if you're using. If you're using AI to acquire skills, then are you. What are you leaving? What are you leaving on the table, journey wise? Like, what is. What is your purpose? And, you know, I mean, it's like you could, like, really go pretty far with that, but, but I do think there's going to be sort of this, like, counterbalance to it going too far. Also, I think the data center thing is really fascinating as well. So there's a city right outside of LA that actually voted no data centers in our city. Um, and so that I think is really fascinating too, because I don't really know how much people know about the environmental impact of these data centers. Like, what it takes to run one. I live in San Francisco and we were having a conversation with somebody last week of like, we're like in the AI bubble. In the AI bubble. So, like, what we think people think and talk about AI all the time is probably not accurate because that's what we do all the time. Um, so, you know, I don't know. I think there is going to be this, like, element of pushback and counterculture that'll be really interesting to see how it plays out.

Speaker C: Uh, I'll have to go. I just made a note to go watch that.

Speaker A: Uh, it's great. It's only 20 minutes long.

Speaker C: I saw the one where Eric Schmidt got got booed off.

Speaker A: Yes.

Speaker C: That one was like, um. And he tried to fight it and it was like. It was like. It was exactly like jumping into the fray on Reddit. That did not go well for him.

Speaker A: Yeah, no, thank you. Our pass.

Speaker C: Well, awesome. Well, how can, uh, if folks, uh, really enjoyed the conversation and want to learn more. Do you publish on LinkedIn? Are you big on TikTok? Where do you. Where do you spend your time as an online person?

Speaker A: I try to stay off TikTok as much as I can.

Speaker C: Um, so it's just Jennifer Peters and then I found, uh, you know, throwing a little Ollie in there. Got me there very quickly.

Speaker A: Yeah. If you look at Jennifer Peters and Ollie on LinkedIn, I'm there. There's probably some stuff out there on YouTube. Who knows how old it is, but yeah, LinkedIn is probably the best place to find me.

Speaker C: Awesome. Jennifer, really appreciate it. We'll have to have you back in a year and see what other kind of magical things you figured out.

Speaker A: See what kind of weird stuff happens.

Speaker C: Uh, awesome. Thanks for coming on.

Speaker A: Thank you.

Speaker B: Rate and share to stay up to date with retailgenic, your guide to the future of AI shopping agents.