Lovisa: AI, Agentic Shopping & Lean Ecommerce Ops

[00:00:00] Neil Du Plessis: Our brains are wired to prefer simplicity over complexity. That's just kinda how how your brain functions. And I would argue that the traditional commerce website, the way that we know it today, is not the most simple way to interact.

[00:00:21] Kailin Noivo: Hey, folks. Super excited for our chat here with Neil today. We tackle a lot of different topics, mainly AI, but two different type of AI topics. The first one being, hey. Is traffic patterns gonna change over time because of agentic search? So that's the first one. And then above and beyond that, we tackle how to actually operationalize AI in lean teams. So really excited for today's episode. Welcome to another episode of the ecommerce toolbox experts perspective. Joining us today, we have Neil Du Plessis joining us all the way from the West Coast. So really excited to have him on the show today.

[00:00:54] Neil Du Plessis: Hey. What's up, Kailin? Thanks for having me. I appreciate it.

[00:00:57] Kailin Noivo: Awesome. Well, Neil and I have known each other for probably six years now, all the way from your guest time to now. So maybe talk to us a bit about your career journey and how you ended up in your role, and you're actually kind of transitioning right now. So, yeah, talk to us a bit about that.

[00:01:09] Neil Du Plessis: Sure thing. Yeah. Look. I spent the majority of my career in the digital side of the world. Most recently, I was vice president of global ecommerce at Lovisa, a fast fashion global jewelry company. Prior to that, I'd spent some time in the software developing world at Fiserv and Clover. Did a short stint in kind of b to b as well on the still on the ecommerce side, but it's really been all digital and ecom going back to the days of Neon. I met, as you mentioned, when I was heading up operations for Guess Stores America.

[00:01:40] Kailin Noivo: Very cool. So let's maybe hop right into it. You and I have had a couple of off camera combos about AI and ecom. So let's maybe break this combo up into two categories. So let's maybe look at it from an operational level, and then we'll talk about the fun stuff, which is kind of like agentic shopping and business storefront change and all of that after. So, yeah, maybe kick us off with from an operations level. Like, how are you thinking about it from customer service to fulfillment to website automation and code generation and all of that? So very broad question, but maybe kick us off there with your thoughts. Yeah.

[00:02:15] Neil Du Plessis: I mean, look. It's everywhere at this point. You know, even at Lovisa, we had made sure to start including or finding ways to pull AI into our workflows, whether it'd be for efficiency, productivity, or just capability, really. So in a business like that that has a relatively high SKU count, you gotta think things like photography and product copywriting. You know, you're talking about a close to 7,000 SKU product assortment with anywhere from a 100 to two fifty new SKUs per week, which requires a lot of turnaround, right, in terms of writing that product marketing copy, getting flat light shot, getting on body shot. So from that point of view, AI has been very, very helpful from a productivity point of view. Because think about how many times can you put an earring to a model's ear in a day before it starts to turn red and it's uncomfortable and it gets swollen, and it's to your limit on that front. But you can obviously keep shooting on a silicone ear as much as you have time in the day, and then that, after the fact, gets placed onto onto a model with the power of AI. Same with the copywriting. Right? You can kind of set and forget, create your brand tone of voice. So that was pretty big. Obviously, outside of something like Lovisa, just as far as web dev is concerned, I mean, vibe coding has kind of become the hot topic or still is a hot topic, I suppose. I've used it myself. I've created some products. I'm sure you've dabbled with it. Even my mom has dabbled with it. She's built something that she sent me the other day, which is, like, this is amazing, crazy.

[00:03:46] Kailin Noivo: I love that. I love that.

[00:03:47] Neil Du Plessis: That somebody is doing that, and it's you know? But that's the world that we live in now, which is phenomenal. Anybody that has an idea can sit down in front of a computer and now just bring that to life, which is I mean, I love that. Absolutely love it.

[00:04:01] Kailin Noivo: It's funny. As you know, I was with, uh, some family in LA last week, and my niece, who's 11, built a website called seacreaturekingdom.com, so you should check it out. And she loves animals, and she completely, like, vibe coded it out and even built a gift shop. So it was really cool to see it. And she's 11, and she was, like, deploying on Vercel and Netlify. And I'm like, what do you do? Like, I was like, when I was your age, I was playing with Pokemon, so it was pretty insane to see. But I love how it's been democratizing everything. What's been the single best use case as an operator that you've seen? And was it the product descriptions? Was it really the imaging? And I know every business will be different for you guys, obviously, at high SKU count, but, yeah, I'm curious.

[00:04:38] Neil Du Plessis: Yeah. I think for me, it was, you know, those are two big operational kind of lifts, right, that need to be dealt with just as kind of day to day BAU type work. But then I think access too. Right? We access to kind of data analysis is another big part of this. We've had, obviously, crazy advanced analytics tools for many years that you can do a lot with. But, again, it's I think it's that power and the ability, and you see it even with Shopify now. Right? Like, they've added the same ability to kind of query through their own agent in your Shopify back end. I'm not super familiar with Salesforce that they have something similar, but I wouldn't be surprised. So you see it starting to enter that world as well, and that was super helpful because I'm like, I'm not even that old, but, wow, I'm that old school person who is, like, diehard Excel. Give me Excel, and I'll build anything in there. I'll analyze. But, obviously, now you can benefit from the power of what an agent or an AI can do in seconds that might have taken you an hour or two.

[00:05:41] Kailin Noivo: Yeah. It's pretty powerful. Obviously, we're on the BI side. And, candidly, what we're seeing is, previously, you needed teams of five, ten analysts to look at BI data. You needed, like, five to 10 engineers and a central IT team to run, like, a global operation. And now we're seeing really, honestly, technology democratize that type of insights and value to smaller teams. And when you kind of started working with Noibu, we really only focused on mid market and enterprise businesses, and now we've been welcoming a lot of SMBs. They use the product differently. Right? They don't have time to log into the dashboard and poke around. They don't have, like, a dedicated afternoon to do that. They more so want the AI to leverage the data and surface them the insights. So we've been seeing a huge amount of adoption there, and I think that's kind of the future of that category. So, then, what you're saying makes a lot of sense. Well, let's dive into a bit more of a controversial topic. Right? I think everyone's bought in to operations, use cases, and things like that. They've seen the value. How is AI and agentic shopping in your opinion gonna potentially change customer behavior, if at all?

[00:06:49] Neil Du Plessis: I think the customer behavior is already changing, and it's not something that we can prevent. I think our brains are wired to prefer simplicity over complexity. That's just kinda how your brain functions. And I would argue that the traditional commerce website, the way that we know it today, is not the most simple way to interact. You've got a billion buttons to click on or screens to swipe through to go through checkout. Even accelerated checkout is still kinda clunky sometimes. Shop Pay, Apple Pay. Apple Pay and Google are probably kind of up there as far as ease of use is concerned. But I guess the kind of pie in the sky of you here is that multimodal agentic interface where, hey. I'm looking for this. Great. Here are some options. Great. Thanks. Buy it. Send it to my house. And it just does that. Right? I think the argument is, well, you know, it's not nearly as rich of an experience if you're in a chat interface today compared to a traditional ecom site. And, of course, that's true. But like you were saying earlier, the traditional ecom site had had many decades to evolve and mature, improve, and kind of try and and cater to the different user behaviors and wants and needs. Whereas, the agent is an agent interface. It's still in its kind of nascent stages, so I think that'll probably also become kind of a richer experience over time. And I think one of the big driving factors behind this is, and we didn't necessarily cover this earlier, but but it's obviously well documented, the amount of capital investment that's going into AI. The big question, right, that is where is the revenue generation gonna come from? And I don't think it's necessarily just gonna they're not gonna cover that level of capital investment with advertising. I think commerce probably is one of the revenue streams where you will see that kind of start to drive this conversation as well where, of course, there's money to be made in processing transactions, and that could be something that I think could be quite interesting there.

[00:08:52] Midroll: If you're listening to the ecommerce toolbox, you're entitled to a podcast exclusive website audit. Go to noibu.com/podcast-audit for a free scan that uncovers the hidden friction blocking your conversions and shows you where you're leaking revenue.

[00:09:07] Kailin Noivo: What percentage of transactions do you think will migrate from a browser to originated out of an LLM? And what's the timeline?

[00:09:17] Neil Du Plessis: If we look at the case study, the recent one with ChatGPT and Walmart and Etsy, I mean, a massive failure. Like, the general consensus was, um, didn't work. I think the Etsy one is still live. I actually tried it out a couple weeks ago. The experience is fine. You know, it's not the best in the world. But with that kind of framing, I think timeline is not short — twenty four, thirty six months, maybe, where you start to see more of a real thing because we're talking about — that is something that is unlikely to happen. But what percentage of? Eventually, all of it is my guess. I think that's the controversy around this, right, is because the argument is, well, brands then lose that owned moment. Right? The moment where the brand is in control because you're on their right property, they get to control that entire experience. I would argue that that experience only existed because that is what we have. Right? Like, we've gone down this path, the evolution of the traditional commerce website. The point where we're at today is obviously a byproduct of the last forty odd years worth of development and evolution, like we said earlier. But I think we've plateaued. We were kinda saying earlier that we're we're maybe at a point where that experience is about as good as it could possibly be. And the next step is, okay, multimodal interface. Google, actually, I mean, the signs are kind of there. The fact that they released the universal cart not too long ago, I think is a really early indicator maybe of them kind of trying to lock in early here, because they must also be cognizant and aware of the fact that they're gonna lose out on ad revenue as the traditional Google search becomes less of what was in its prime because that traffic has started to shift away. Moved to OpenAI. It's moved to Anthropic. It's moved to Perplexity. It's you know, there's all these other options out there now, and I think Google is probably aware of that. So I think they're locking in early with something like universal cart, but I wouldn't be surprised if you start to see similar types of moves from some of the other AI companies.

[00:11:25] Kailin Noivo: I'll take the other side of the argument just to make it fun. I think that unless there's a step function change in the actual experience, I think it's gonna, like, flatten out as, like, a channel. Is it it'll just be a channel where, like, you'll get traffic for, like, random kind of very bespoke queries that are, like, hyper personalized to to you? I don't know. I think even with my own behavior, and, like, it's been really buggy trying to shop for ChatGPT. I've never tried through Claude. So, yeah, I think it mustered a big change. It's gonna be I think it's gonna be a small channel. But that said, it's gonna be interesting because we don't know what we don't know, and there may be something that comes out. Right? Whether it's a wearable device from OpenAI or something like that that actually could change the game, where, like, I don't know, you're walking with glasses on, you see someone walking by in an outfit, and you're like, oh, I wanna buy that. And that's just a random example. But if there's, like, a few of those that come out, I can see it chipping, but we track the data right now. And to your point, it's kinda like it's pretty underwhelming. Just traffic, conversion. Yeah. It's kinda like, it's just not in the mood yet. But, yeah, I'm also curious. I feel like online shopping at the beginning sucked too. You needed to, like, send the money order or whatever, and it grew over time. But, yeah, I don't know. It'll be interesting.

[00:12:43] Neil Du Plessis: It is. I think the futurist in me will come out. Like, if you think about it, you've got a connected TV, for example, today, right, where a lot of people have connected TVs. It's not impossible because I think the technology exists. Right? I don't think this is a question of can it physically be done. If it can physically be done, it's in the Walmart examples of proof of it. They're not good examples, but it can be done. But, I mean, think of a world where you're sitting on your couch with your connected TV. You've got maybe an Alexa or Google speaker somewhere or whatever the case is. And to your point, you maybe see something, you go, oh, I want that. Your connected device is listening to you. You know, you engage with a user of your connected TV now to show you the rich experience that we're kind of clamoring for here from a shopping point of view. Quickly switches over to that view. You engage using your voice. You get to see examples. You get to see video, and you purchase. That's it. That experience is a lot richer and a lot easier than it is to pull out your phone, go to a website, go to a URL. And it arguably, I think this idea of, well, you know, discovery and this and that, discovery happens off-site anyway and has for a long time. That's not really where organic discovery happens anymore. So I think the big the while we're on the topic of kind of things that aren't really yet pressing to throw at you there is, when does the first truly kind of, let's call it, born digital brand launch? Brand that leans into this idea and says, we don't have a traditional ecommerce site. They might have some kind of a website or because I think we're at a place where you still need information to pass from brand to AI. Right? There's all kinds of operational info that needs to be passed along. But forgetting that for a second, when do we see that first brand that says, we tell the story, the brand gets built across external social channels and other media channels like most brands do today, but the transaction itself switches to one that happens through some type of agent interface instead?

[00:14:48] Kailin Noivo: And I could see that changing too as, like, the personal agent becomes more popular. Like, if you have, like, a family agent or something like that, I could see that changing, especially for, like, commodity goods, groceries, stuff like that. So it's interesting. Neil, as we look to wrap up, maybe your top prediction for 2026 for commerce, top trend?

[00:15:11] Neil Du Plessis: Interesting. I think there's obviously, there's a lot of things happening, but I think the emergence of leveraging the power of AI to, again, make that shopping experience based with the current kind of way that we transact and how we conduct commerce a richer experience through AI, everything from virtual try on to some of the same things that, again, Google have recently released with kind of their version of shopping and virtual try on. Personally, at Lovisa, we saw a lot of success with that and partnering with some really great vendor partners on that because it, yeah, it's that you know, we were talking earlier about what's that upper kind of limit of how good can the traditional web experience be. That kinda takes it just a little step further, bringing that digital thing more to your real life. But, yeah, I think we'll see some more of that happening in the next couple months here and kind of be adopted. But I honestly, the pace at which things are changing, I think it's kind of like throwing a dart at a dartboard in a dark room here. No idea. Did things are your niece vibe coded something at the age of 11. I mean, we used to say there's not a 15 year old here that's gonna come out with something in the next five weeks that everybody's gonna go, you know, mind blown. But that's the great thing about it. I think that's the great thing about the age that we live in at best.

[00:16:30] Kailin Noivo: I will see you, Neil. And thanks for taking the time to hop on in. It was great. Absolutely. Really appreciate your thoughts and insights.

[00:16:36] Neil Du Plessis: Yeah. Thanks so much. I appreciate you having me.

[00:16:39] Outro: The ecommerce toolbox AI and retail is brought to you by Noibu. To find out more about Noibu and how we unify error monitoring, site performance, and experience analytics to uncover growth opportunities and skyrocket your revenue, visit www.noibu.com. That's noibu.com. And then make sure to search for the ecommerce toolbox AI in retail on Apple Podcast, Spotify, or anywhere else podcasts are found, and click subscribe so you don't miss out on any future episodes. On behalf of the team here at Noibu, thanks for listening.