Retailgentic, presented by the ReFiBuy Media Network and hosted by Scot Wingo, is where Agentic Commerce meets retail innovation. We help retailers and brands prepare for the future of agent-driven shopping with actionable news, expert insights, dxis, forward-looking predictions, and practical guidance on Agentic Commerce Optimization (ACO).
Welcome to the Retailgentic podcast where we explore the intersection of retail, ecommerce, and AI agents, which we call Retailgentic. Our mission is simple. We are shining a bright light on the new agentic AI shopping megatrend, which will disrupt the trillion dollar digital retail market. The stakes are high. Google, OpenAI, Microsoft, Perplexity, Amazon, Apple, Meta, and Anthropic are all fighting to be the consumer AI shopping agent winner.
Speaker 1:There are gonna 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?
Speaker 1: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 Retailgentic, your podcast partner in navigating our AI shopping agent future. And now here's your host, Wingo.
Speaker 2:Welcome to this episode of Retailgentic. We are coming to you live from Retailclub in Huntington Beach, California, and we have a live audience. Hello, live audience. Alright. Joining us today is a very sneezy Jason Goldberg.
Speaker 2:He's going be helping me co host this one. And then Ekta Chopra, she is the Chief Technology and AI Officer at e. F. She also has an awesome substack called AI Chef, and she's joining us hot off the main stage where she dropped some wisdom.
Speaker 3:Yeah, thank you.
Speaker 2:Yeah, you could tell it was popular because as she worked her way over here, there was was a cluster of 30 people chasing her down entourage. So you're very popular. We really appreciate you taking time to come speak and to we discovered this is our one year reunion. We had you on the Jason and Scot That show a year was a very popular episode. What we learned in that episode is you're able to see into the future.
Speaker 2:Don't know if it's, back then it was maybe four or five months. I don't know if that's gotten longer or shorter because AI has sped up.
Speaker 3:Man, phase of change is different every The
Speaker 2:amount of things that have happened in the last year, your title, you were Chief Digital Officer back then, and now you've got a combination of CTO and AI, which is pretty cool.
Speaker 3:You.
Speaker 2:Congrats on that.
Speaker 4:Thank you.
Speaker 2:And then e. F. Has continued to grow and do amazing things. The audience that may not know about e. L.
Speaker 2:Give us a little update on e. L. F. And then talk about your career and then we'll jump into it.
Speaker 3:Yeah. So e. L. F. Beauty, we're a house of brands, we have the e.
Speaker 3:L. F. Brands, which is e. L. F.
Speaker 3:Cosmetics, e. L. F. Hair, e. L.
Speaker 3:F. Skin. Then we have Road, which is the Hailey Bieber brand, Naturium, and Well People. So it's a global company, 30 consecutive growth quarters, which I'm so proud of.
Speaker 4:High five on that.
Speaker 2:Yeah. Good job.
Speaker 3:Elf is a bold disruptor with a kind heart. And I think that's sort of our hallmark.
Speaker 2:Yeah. And fun fact, Elf stands for eyes, lips, face.
Speaker 3:That is correct. Yep. Somebody at Jeopardy! Though confused it with ears which I still don't get it.
Speaker 4:Did they get credit or did they? No. No. They got buzzed.
Speaker 2:Yeah. Not buzzed. I was like, what?
Speaker 3:Yeah. But it became a fun meme and a social campaign
Speaker 4:which
Speaker 3:was very
Speaker 4:That's kinda... You know, you made it win your Jeopardy question.
Speaker 3:A 100%. A 100%.
Speaker 2:And then what does your role entail at e. F?
Speaker 3:So I oversee all aspects of technology, which includes, you know, the plumbing, which is oftentimes the infrastructure, but then data, marketing technology, AI, and of course, and security, which is really important in this world as well.
Speaker 2:Yeah. Yeah. So you were on the main stage. What were what were the big topics du jour on the main stage for you today?
Speaker 3:I think people are still trying to figure out, you know, sort of how do you really get value? And where do you get stuck when you're trying to do the AI program. And it boils down to sort of, you know, there's two schools of top thoughts. One is let it rip or lock it down, which is I'm gonna give AI to everybody or I'm gonna lock it down. And for us, we've tried to do both where we don't kill innovation.
Speaker 3:So I think the topic is how do you do that at a scale and different organizations are going through different, you know, issues. And I think it boils down to, does your leadership believe in it, your org structure? There's so many nuances.
Speaker 2:Yeah. Yeah. We deal with a lot of customers and some of them will have a SWAT team that kind of sits over here.
Speaker 4:Mhmm.
Speaker 2:And they're gonna go really fast, but then assimilate a lot of the things they learn into the org. Sometimes they stay separate and maybe they'll get absorbed down the road. You apparently talked about some pod thing that was very popular that generated a little bit of buzz. Yeah. What are pods?
Speaker 2:Are they in the basement? Should we be concerned?
Speaker 3:They are... I wanna say, ultimately, it's the role of a product manager who really understands the domain. So we have three pods. One is focused on agentic commerce or agentic growth. The other one is sort of intelligence and intelligent operations.
Speaker 3:And then third is human agent workforce. And I think all three of them basically have everything that you need to truly do something at scale. So including the work around data, change management, integrations, and training and education. So it's important that, subject matter expertise tied with enablement is how I think about the pods. But at the end of the day, it really is no different than a product manager role who understands the subject matter expertise and is bringing new products to life.
Speaker 3:Yeah. You know?
Speaker 2:Let's dig into the agentic commerce one. Yeah. Is this people's full time job or they get kind of tagged to also be on this team in kind of like a dotted line type situation?
Speaker 3:Yeah. So like right now it has been sort of whoever has the time and expertise, but now we've just opened up these positions because they're gonna be dedicated pods for all our brands essentially because it does require that level of focus. So through our process, obviously, I tie it back, there's the product work that we have to do to get the product ready to be shown into discoverability, but then there's content that has to be enabled. And what is the context that you have to go after in terms of what are the consumers looking for? And that's a lot of work.
Speaker 3:Then tied back to how do you then plug that into the processes and the different cross functional teams that need to understand and be educated.
Speaker 4:Yeah. I'm I'm always really curious because I feel like one of the unsung challenges of this whole transformation we're all living through is the org change man.
Speaker 3:A 100 Yes.
Speaker 4:And so, obviously, you're you're you're creating full time positions for this pod. You have to hire people. You're not gonna get anyone with a degree in agentic commerce.
Speaker 3:A 100%. Yes.
Speaker 4:And even finding someone with any agentic commerce experience is unlikely. And then in a perfect world, I imagine there's two domains you want them to be expert in. You'd love them to know something about beauty or cosmetics, and you'd love them to know something about agentic commerce. So what what is the profile of some... Like, obviously, you're gonna have to find someone who's...
Speaker 4:Has proven capabilities
Speaker 3:Mhmm.
Speaker 4:And they're gonna have to upscale themselves or you're gonna have to upscale them Yeah. To this role. How are you thinking about that?
Speaker 3:So the non negotiable is they have to... I'll talk about the agentic growth role. They have to know commerce, full stop, period. They have to know... They have to have a passion for it.
Speaker 3:They have to know it. They either worked as a digital product manager and so forth. Second is they have to love AI. They have to love what the technology does. And then the third is the communication and the curiosity and being able to challenge the status quo.
Speaker 3:Because part of the struggle here is people think that you can just plug in things, But no, there is... The reimagination of how things work is just as important.
Speaker 4:Yeah. I I have a pro tip. Here's my new favorite hiring practice. Ask them to open up their email and do a search on you have reached your quota and see how many how many hits they get.
Speaker 3:Good. There you go. I'm gonna try that next time.
Speaker 2:You guys... We mentioned this a year ago that you guys were already starting to invest in this this area.
Speaker 4:Mhmm.
Speaker 2:How do you go to... You're a public company. How do you go to the CEO and the board and say, hey. This is important. We need to invest in it.
Speaker 2:You're gonna ask me the ROI, and it's too early for me to tell you the ROI. But you're just gonna have to trust me that
Speaker 3:Yeah.
Speaker 2:It's gonna be there. How how do you how do you do that?
Speaker 3:You know, in that sense, is very lucky. I think we're... I I feel very grateful. I've been there ten years and nine months that our board and our leadership is so forward thinking. So on our board...
Speaker 3:I mean, November, I'm going back to the board. Every quarter, we talk about the importance of being... Your consumer is changing. Your consumer is on these LLMs, and discoverability is becoming important. So in fact, we're not chasing for funding.
Speaker 3:They're telling us how quickly are you gonna get there.
Speaker 2:So it's pulling you over.
Speaker 3:So it's so it's really interesting that I feel proud that we have a board that is very forward thinking. And, of course, we also have to back it up with receipts, we have to back it up with, all right, we are not wasting money. We are actually seeing the momentum, and we're actually seeing that this is making an impact.
Speaker 2:Yeah. And there's early on, it may not be dollars, but there can be x number of products have been upgraded to be agentic friendly. Or a lot of people talk about AI visibility, which is more of a sentiment type of a thing, that kind of thing. So there are measurements, but there has to be a little bit of faith that those dots are going to connect to
Speaker 3:the Yeah, I I think that's been the challenge because we talk a lot about measurement. Frankly, anybody that tells you they have figured it out, haven't figured it out. So I think it's more of the way I think about it is readiness. There is a fact that you'll start seeing that. I think there was a 60% of consumers are now not searching as a traditional search, but they're actually going to chat GPT and they rely more on chat GPT answers than they do on anything else.
Speaker 3:Right? So I think if you just look at that number, then I think you could probably... The way we've looked at it is more of, hey, are we even visit... Are we even visible there for our top product? So we've been able to break down, you know, where do we have the right to win?
Speaker 3:Because you can't tackle everything, but where you have the right to win, where the conversation is happening, let's go do that work and get ourselves ready. So we have a very focused approach. We're not trying to boil the ocean, but we are definitely looking at areas where we have the right to win, and we should be winning.
Speaker 2:Yeah. It's good you have a board that's proactive because when... The most frequent call we get at refi buy is urgent from a retailer or brand, and it's usually a pretty senior person, like a Chief Digital Officer. And when I ask them, what inspired you to call all of a sudden? They're like, well, someone on my board was looking for a product or my CEO.
Speaker 3:I couldn't
Speaker 2:find it. Then I searched and you guys seem to be the only ones that can figure out how to connect all these dots. It's kind of funny that they're much more reactive than being proactive.
Speaker 4:I mean, it's funny and it's not funny. I feel like echoing your experience, every single client now, there's a board member that did some superficial query search and sent a... An uncomfortable question mark email to the
Speaker 3:Where are you?
Speaker 4:Leader. And then I'll tell you the thing that compounds that you mentioned still the Wild West of of visibility measurement. Yeah. There's a thousand vendors out there that that have some concept of a solution. Yeah.
Speaker 4:And every one of them is now running an audit on your brand. Yep. And they're sending that audit to one of your board members. Yeah. So after you get that, I ran my own search.
Speaker 4:Now you're getting all these people are coming out of the woodworks and saying you're doing well or not doing well.
Speaker 3:And I think part of the challenge also is that because there's no measurement and everybody's doing their own thing, it's about I asked these 10 questions and you didn't show up on any. Well, were they even relevant to us? Were they part of the context that our consumers were looking for? So there is a lot of noise because everybody's coming out with their own set of questions essentially. Yeah.
Speaker 4:There is something I'm mildly excited about. Almost all the measurement we've talked about so far is what we would call synthetic prompting.
Speaker 3:Yes.
Speaker 4:Someone, either a robot or a human, invented a prompt and then tried to measure how that prompt showed up in in the multitude of AI services. Yeah. And that's that's a useful tactic in the ecosystem. What none of those tell you is what real humans actually typed into the AI surfaces.
Speaker 3:Exactly.
Speaker 4:And something I'm mildly excited about, Google made an announcement last week that in their Merchant Center, now have a tab called Agantic Insights.
Speaker 3:Yep.
Speaker 4:And they are now for the first time sharing real user data, at least on Google surfaces
Speaker 3:Which is fantastic. About
Speaker 4:what prompts. So this is, to me, the first validation of what things are people really typing Exactly. That impact
Speaker 3:Exactly. Yeah.
Speaker 2:Yeah. Since this is one of the sessions that has an audience, any audience questions for Ekta? Let me restate the question. The question was, how do you feel about build versus buy? The classic.
Speaker 3:Yeah. It's
Speaker 2:an oldie but goodie.
Speaker 3:Yeah. I mean, I think in general, it's been more buy for us, but what we're finding is for certain things that we are building because technology is the same everywhere now. It's the uniqueness of what is it that you wanna differentiate yourself with. So I do think that there's a little bit of aperture opening up for building, but majority, I've been just buying, you know, sort of the solutions. But I do think that there are things that are coming together as more proprietary, things that I wanna own my data.
Speaker 3:Like, that is where we're investing in building.
Speaker 4:Ekta is lucky that she's in a company that that sort of lean forward. But in a lot of companies where they maybe didn't lean forward, it's hard to know where to start. There's all these multitude of of people that are bombarding at you. And what Yeah. How how do you get started?
Speaker 3:Yeah. So what I always find is sometimes you can be the only voice in the room and nobody wants to listen because they're like, oh my god, there's so much to do and blah blah blah. So I always find it helpful when I bring external voices. So the first thing we did was when this... I think this was a year and maybe two months ago before before we met.
Speaker 3:We actually brought in an external expert in search and how search was evolving and so forth. And that was like a kind of a fire in the belly. Oh my god. Are we gonna be so behind? And then that sort of helped us gain momentum.
Speaker 3:Alright. Content is one thing. What about your product feed? What are you gonna do there? And luckily, we met.
Speaker 3:And then, you know, the... So I think it is a... For for me, what's been always helpful is any new topic is scary, and it's a big change initiative because we've had to educate our, you know, sort of our regulatory teams because skincare, it's so... Claims and all of that stuff. So I think it's been a lot of education.
Speaker 3:Frankly, the education has taken longer than the technology was able to catch up to it. So I think it's external voices can sometimes really help because they're not just the internal noise.
Speaker 4:Let me ask a follow-up question. I think that's that's great advice both sort of the external validation but also kind of leaning into the fear uncertainty and doubt a little bit like can can sometimes be helpful to keep up with the Joneses. But once you've done that... So now... Okay.
Speaker 4:Cool. We weren't paying attention. Now we are. Where should we invest our limited resources to get started? Like, did you have a hypothesis that like, is the data the best first place?
Speaker 4:Like, why didn't you start with a ChatGPT app or something?
Speaker 3:Yeah. Yeah. I think it goes down to... I'm a strong believer that you have to tackle the foundation first. So, yes, you can go to the...
Speaker 3:You know, at the surface level, bad data is equal to bad AI. So if you don't solve for the plumbing issue first, I don't think you can scale. So I'll always look for... There's traditional platforms that are still catching up and they've added AI into their, you know, name or they're now claiming to do certain things, but then
Speaker 4:There's there are no companies that just added AI. That's crazy. But
Speaker 3:then there's the one who were actually born in this era. So how do you sort of... So I think it it is a little bit of a dance that you're
Speaker 4:Scot and I have met several businesses at this show that have been in LLM AI since 2006.
Speaker 3:Yeah. Yeah. That's 2006. There you go.
Speaker 2:Yeah. Or like, you mean '26? No. No. Twenty years ago.
Speaker 3:Yep. Twenty years ago.
Speaker 2:So a lot of rightful focus on discovery and consumer experiences. How are you thinking about back office applications, you know, inventory management, fulfillment, that sort of stuff?
Speaker 3:Oh, yeah. Definitely. You know, lot of people talk about marketing and they talk about, like, the value this will bring from a revenue perspective. I'm a strong believer. Supply chain is where there's a lot to be unlocked.
Speaker 3:So we actually are already partnering with some really innovative company. One of them is here, Lyric AI, which we've invested in for forecasting and demand and planning and so forth, which has been just incredible. And what's incredible is that our traditional planners and demand planners and supply planners are actually now engineers because technology is the easy part. So it's where I get very excited is when the teams are not stuck in the containers of the past, as would say it. They are actually very forward thinking.
Speaker 3:So I truly believe technology is the easy part. It's harder part getting teams to think differently now. You know?
Speaker 4:I think you said something that I tell people a lot. I just wanna validate that I heard you right. But the discovery and customer facing surfaces are going to be monumentally important in all of this, and they're our future. If you need a favorable economic ROI this quarter or this year, those those... If it...
Speaker 4:Those AI use cases around efficiency and back of house are likely gonna have a faster return.
Speaker 3:A 100%. More than that, I would say is there is this fear. If you can give a governed way for your teams to, you know, understand AI and use it, I'm finding that AP, AR, payroll, the inventory clerks are just figuring out how to make their work easy, faster, and actually build this, you know, boots on the ground momentum that I've never seen before. So...
Speaker 2:Yeah. One question that's very topical here at the show is we just had Muse launched an agent. There's Instinct. There's these new category we had answer engines. And then we had the the kind of the vertical Yeah.
Speaker 2:Engines. And then some of them were just like focused on fashion, and others lived at like vertical shopper
Speaker 3:Yeah.
Speaker 2:Agents like Rufus slash AFS and Sparky. Now we have this other category, which is these personal assistants.
Speaker 3:Mhmm.
Speaker 2:Do you have a point of view as Elf opening up the doors? Are you kinda keeping an eye on these? How do you
Speaker 3:feel about always open for test and learn. Always. Having said that, like, of course, governance is important to us and so forth, but we are already gonna do... We're doing testing and learning there. So I think, you know, along with the testing and learning, you also have to take the learning back to what do you need to build in your foundation that'll help you scale it.
Speaker 3:If it works and you wanna lean in, you still have to go back and solve for the foundation.
Speaker 2:Got it. Any other audience questions? One question I had is you you have all these brands and then you you... Your team has to service these brands.
Speaker 3:Yeah.
Speaker 2:Do you act kinda like an internal agency? Do the brands kinda clamor like... How do you... There's probably four different voices there asking for things. How do you how do you navigate all that?
Speaker 3:Yeah. I mean, I think we are very conscious about the fact that we don't wanna kill the brands running and innovating. But when I think about the layer of governance and standardization, it has to come through the the corporate team in many ways, specifically when it touches AI and payments and some of the stuff that, you know, governance is required. But that is driven by the internal
Speaker 2:Yeah.
Speaker 3:Internal team.
Speaker 2:Have you guys... So you're you're in an interesting position because you have a a vibrant direct business, but then you also sell through a lot of retail partners. Yeah. Have you partnered with your retail partners to take a lot of learnings you've learned on the direct side and try to... Let's say you discover a road product needs a certain piece of data in it for a specific reason.
Speaker 2:Are you able to get those folks downstream to think about putting that in? Or is We that
Speaker 3:we are. And I think with some retailers are... It... It's easier than others, but we are definitely on Amazon. We're doing a lot of work.
Speaker 3:You know? On Walmart, we're doing a lot of work. Target as well. So I think it is a journey, though, because what they allow and so forth, bigger organizations, multiple layers versus us trying to move quickly. But we always try to work with our retail partners.
Speaker 4:It fair to say I feel like there's an opportunity for those wholesale partners, those retailers, to get outsized returns right now if they're willing to lean in. I feel like there's a lot of brands like you that are open to test and learn. And if you find a receptive retailer, they might get more of your bandwidth than
Speaker 3:A someone who's 100%. More collaboration the better. I
Speaker 4:want to go back slightly. You talked about the importance of governance and the balance. One of the things I run into a lot is traditional governance models they wanted to know that everything was absolutely safe. Like oh is this brand safe? Could someone ever do something that's erosive to our brand?
Speaker 4:And my feeling about our current state is that's no longer a reasonable goal or expectation. You can't do something that's perfectly brand safe or guaranteed to be brand safe. So are you finding that your governance partners are evolving with the pace of change and all of these new evolutions?
Speaker 3:So it it did take regulatory about nine months to understand and, like, really... But they are my best partners right now, and they're actually investing their time. And and I think that's something that I'm very proud of, at least, f, because our teams definitely... They wanna do the right thing, but then they also wanna, you know, continue to push themselves. So
Speaker 4:Yeah. I feel like an irony that I like to point out to some of them when I'm in a a pesky mood is that they're gonna get replaced by the robots before we do.
Speaker 3:Might be some truth to that, but yeah.
Speaker 2:Yeah. I can never get them to understand that they're like, well, could someone ask an AI, so we're seeing this metadata. They're like, can someone ask the AI and get the data? I'm like, yeah, but they can do that on Google or any like, all this stuff's out everywhere. Like they're just so obsessed with the AI thing that they don't, sometimes logic doesn't
Speaker 4:In the sign early days of social, literally had like a CEO of a retailer tell me we're not doing any of those socials because people would say bad things about me. And I'm like do you think if you don't have a Facebook page, that's going to stop people from saying bad things about you?
Speaker 2:Let me introduce you to the site called Reddit. Yeah. Be careful. So let's think forward a little bit. So let's say we're here next year.
Speaker 2:What are some of the things you think are going to be happening over this next twelve months?
Speaker 3:I mean, I think with the rise of Muse and so forth, so the agent shopping, I think Meta, I will give them a lot of credit for sort of... In this area, at least, I feel like they are going to win the race from my perspective. Maybe that's my prediction. But what does that mean translating to the brands? I think those who are waiting to sort of implement agent identity, they should do that sooner rather than later.
Speaker 3:Because if the shopping increases where agents are shopping, you better know who that agent is that's coming and how does that tie to the customer profile. So I think that's a little bit of the plumbing work. Just like the CDP work that people needed to do for personalization or CRM, I think for agent relationship management, I think you have to start thinking about that.
Speaker 2:Yeah. Do you think you think better we'll tie this into Instagram in some interesting way? Possibly. I'm sure that's
Speaker 3:a huge
Speaker 4:channel for you guys.
Speaker 3:Possibly. Yes. I I believe so. Yeah.
Speaker 2:Yeah. Yes. We're interested to see what they
Speaker 3:do with it.
Speaker 2:Yeah. I have a 19 year old daughter. She's a huge fan. And she... All she had to do is personalize her little Muse agent, and now she's, like, in love with it.
Speaker 2:Yeah. She'll never turn that thing off. Yeah. So it's very...
Speaker 3:It is kinda cute. It is kinda cute.
Speaker 2:I saw that they now have them where they're animated and can talk to you.
Speaker 3:Yeah. I don't know
Speaker 2:if you've seen that video or not, but that's gonna... It's... They're leaning into the cute on that. Yeah. Which I think will impact your your younger customers a lot.
Speaker 4:Yep.
Speaker 2:Well, Ekta, I know you're super busy and you're jetting off to the next big thing. We really appreciate you coming.
Speaker 3:Thank you for having Yeah.
Speaker 2:Let's give it up everybody for Ekta Thank
Speaker 3:you.
Speaker 2:Thank you.
Speaker 1:Follow, rate, and share to stay up to date with Retailgentic, your guide to the future of AI shopping agents.