[00:00:00] Welcome to Unpacking the Digital Shelf, where industry leaders share insights, strategies, and stories to help brands win in the ever-changing world of commerce. Hey everyone, Peter Crosby here from the Digital Shelf Institute.
[00:00:26] EyeBuyDirect sells millions of pairs of glasses every year on their website, which makes every instant a shopper spends on that site a moment of truth. Sunny Jiang, CEO and President at EyeBuyDirect Inc., saw the potential of AI to use data to bring the insights of those millions of consumers to actionable life for every corner of the site and for every employee at EBD.
[00:00:51] She joins the podcast along with Sonal Gandhi, Chief Content Officer at The Lead, to outline the details of an AI strategy that is laser-focused on taking AI action for measurable results.
[00:01:05] Welcome to the podcast, Sunny and Sonal Gandhi, Chief Content Officer at The Lead, to outline the details of the digital Shelf Institute of Technology and Design for the Digital Shelf Institute of Technology.
[00:01:29] Really, how you're going to infuse AI into your overall, the way you work and connect with consumers and the way you understand them better, etc. So, you know, AI is everywhere, everyone's talking about it, but impactful AI is another thing. And so, it really has to be connected to a why for the consumer, for the company.
[00:01:52] So, I'm just wondering, what is your why behind taking AI and injecting it into how you work in the industry? Yeah. Thank you, Peter, for this point or question. At EyeBuyDirect, as you mentioned, the short name is EBD. So, pretty much everything we do is digitalized because the brand now is almost purely online player.
[00:02:18] With that in mind, we use a leverage AI in many, many places. And it totally makes sense to explain to people and let people understand the rationale or considerations behind. Really, the first why starts with removing the frictions for customers rather than just adding another technology.
[00:02:45] If you think about it, purchasing eyewear online could be complicated, right? So, people have doubts, confusion, etc. So, we are using AI to, for example, help people to scan their prescriptions, to recommend on some personalized frames, to the lens selections on different coatings, steps, etc. And even to the post-purchase experience, etc.
[00:03:15] So, really, everything we are doing is aiming to simplify the decision makings and to streamline the roadmap. Really, that is the first why regarding the simple experience for customers. The second why, I would say, is to empower our people and team rather than replacing them.
[00:03:40] So, we believe that when we encourage and support our teams to use AI widely and thoroughly, it will release some of our time. With this released time, and we hope that we can focus better on more valuable work, such as the creativities, the problem solvings, and the experience improvement, etc.
[00:04:05] And when we empower our team to use AI more efficiently and deeply, and then, you know, the team will be able to create a better product and a service. So, that's about the second why. And another point regarding why is we look at the AI as our organizational capability for the mid-long-term growth of the business.
[00:04:30] So, when our people are using AI to do studies, researches, analytics, reporting, etc., all such things, when they use them really effectively, and we believe it is helping the organization to build up a future-ready organization or culture, right? So, these are the two or three things regarding why, as per to answer your point or question, Peter.
[00:05:01] And how do you think that the internal teams are reacting to that? Like, I totally agree with your point that it's unlocking time and giving them more time back to do less tactical work. Do you feel that there's a lot of excitement? Is there fear? Is there a little bit of both? I'm just curious the reaction to kind of those priorities. I think it's great that you're being so clear about it, but how is the internal organization feeling about that? Yeah. Yeah. I would say the internal, like, feeling of the teams is,
[00:05:31] overall, it's pretty positive. As I said, the overall business is very digitalized. Everything we do is online. So, like, people in general are very acceptive. So, like, right, they're open and curious to learn with AI, isn't it? But, of course, in individual cases, it depends on the people and the teams. So, like, they can be some of the people, like,
[00:05:56] they're really curious and ready to embrace AI into their daily work. Some of them are maybe a little bit conservative and say that, oh, I want to observe first until I have a mature understanding of how much value it can bring to me. Then, you know, like, I can, you know, like, be really aggressive or, like, ready to use it. It really depends, I would say. But, you know, overall, our teams are pretty much active or proactive about AI.
[00:06:25] And I wanted to dig into your third why, which I just want to make sure I understood it, which is about sort of using AI to unlock the power of data that you already have or that's providing additional data. It's just a way to really be able to put data to work. Is that how I, did I understand it correctly? Or is that, did I miss a nuance? Yeah.
[00:06:51] The third point regarding why Peter was talking about are, you know, like, as we said, you know, like the organization building, you know, like to be sure that we're ready for the future development of the business, right? As we just mentioned that when our teams are thoroughly using or like working with AI, and it is not for just one individual case,
[00:07:20] it is, you know, like becoming a culture in a company, right? And then, you know, we know AI is such a big trend for the future, right? We want to make sure that our teams are not falling behind in terms of the innovations, technologies. And like, we are always there, you know, like about the technology innovation. We know how to interact and leverage AI as a team. That's great. I love that point of view.
[00:07:49] And Sunny, let's talk about glasses, right? Like in terms of a purchase, they're a very considered purchase, right? Like it has to look good on your face. You have to like them. It's not just a, hey, I'm going to go to the store and pick up some glasses. There's a lot that goes into kind of picking a pair of glasses and it's very personalized. So how are you seeing consumers interact with AI, especially in your category and with the offerings that you have to choose your glasses? I feel like it's a perfect opportunity for them to use AI to pick out these types of glasses.
[00:08:20] Yeah. So the first observation we like have in recent years is that customers, they are using AI for, to validate their choices or options. So, for example, by the moment when someone interact or engage with our virtual try-on from the website,
[00:08:45] and then, you know, he or she will turn on the camera, will put the frames on their faces, essentially. They want to make sure that the frames look and feel is something they want, they like. And then, you know, when they see the different colors of the lenses in the frames, they want to make sure, oh, you know, like this is the function or the style that I want. As a whole piece, they even want to make sure that it reflects my identity or my individuality, essentially.
[00:09:13] So, really, the first thing is to validate, oh, this is a good choice. This is the best fit for me. The second observation is that we notice that the customers, they are no longer like using AI just as a browsing tool. So, they use it like as for guidance.
[00:09:35] So, if you think about it, when a new visitor lands on a website, right, he's new, he doesn't have much experience to shop, I wear online. And he doesn't want to be overwhelmed with the thousands of product or frames options, right, or even the best sellers.
[00:09:56] They just wanted to receive a step-by-step super clear instructions regarding how to choose the frames to the lenses and to the checkout process, right? And when it's a repeat customer, we learned from the surveys that he already wanted to know, like to have a visibility of his previous interactions with the products and even with the previous purchases.
[00:10:20] And then, you know, he wanted to quickly see some similar options as recommendations, etc. So, like it's about personalization, customization, etc. But, you know, as I mentioned earlier, they don't look at the AI for a purpose just to browse something, but also, you know, like to make decisions and to really meet their goals when they shop from different backgrounds.
[00:10:48] Sort of that digital concierge idea, like an extremely knowledgeable person who also gets what you look like, what you're trying to aspire to, like all of that. Yeah. It really, do you have, and Noah's a perfect fine answer. I'm just wondering if you've seen the sort of, you know, I think of kind of, you were talking about friction earlier. And I was wondering, what are the kinds of metrics that you sort of evaluate friction by?
[00:11:17] Is it time spent on site? Is it, you know, how quickly can they make it to, from, you know, their first look to, have you seen sort of just, without sharing obviously any data you're not comfortable with, but have you been able to see that in the metrics as well as the anecdotal experience? Yeah, of course. Like we, the biggest metric we are tracking is, of course, about the bounce rate and the conversion rate, right?
[00:11:46] So the conversion rate is really the final KPI that we want to understand very consistently daily basis regarding, oh, you know, like out of 100 people who are using a website, how many of them left with something, you know, like with some project. And that is really the final, very low funnel transactions that we're understanding.
[00:12:12] And another thing is regarding the engagement on different sections of the website. So again, we wanted to understand, oh, you know, like if it's a new customer, which sections he or she interact with from the top navigations to the recommendations, like to the assortment display center.
[00:12:34] So like we, overall, you know, as also you mentioned, Peter, that how much time, like one customer spent on a website and even, you know, on different locations or specific positions on the website, how much time he stayed. And also another example is that if it's a new customer, he interacted with us with the chat, right? And because he doesn't want to really learn or browse the website everywhere.
[00:13:04] And he's going to ask a question. And at least in order to understand whether we are successfully removing the frictions for the customer experience, we are measuring by the total interactive time with our customer service representative and also the problem solving rate. Right. When someone walks away, we do our survey. Are you happy or satisfied with the service that we provided?
[00:13:28] All such things are trying to help us to understand, you know, if we are doing a good job about, you know, the simplification or the frictions removals. It's amazing what digitally native companies like yours are able to achieve with that level of sort of focus. You know exactly who your customer, you know how they're going to interact and you're constantly testing that experience and refining. Exactly. Even for the assortment, right?
[00:13:58] For different landing pages, like the product pages, the catalog pages. We are cautious about how many products that we want to display on one row and how many rows that we want people to have from one glance in order to really increase the interactions and the efficiency of the website usage.
[00:14:18] So all of these things that we also try to measure really, you know, like to remove the confusion or the complexities from the website and then to simplify the life for the work for the customers. And so you, you obviously talk to so many brands and I'm sure the topic of AI every once in a while comes up in these conversations.
[00:14:42] And I'm wondering sort of on the broader, on the, on the broader lens that you have at the lead, how are you seeing other brands adjust to this, to the consumer engagement? And how many are leaning in as sort of thoughtfully and aggressively as, as Sunny is at, at EBD? So I think that responding to how consumers are searching for everyday things is, is imperative for every brand, right?
[00:15:09] So they are, most of the consumers are now going, you know, changing the way that they search for things. So if you're buying an object, let's say you want to buy a graduation dress for your, you know, child's graduation ceremony. You would put in graduation desk, a dress in a search engine and go from there. But today you're really searching in a more comprehensive way. Give me ideas for a graduation desk that I can wear in this weather at that college, in this color, this length.
[00:15:39] So you're really like, this is how we search now. And some of the, so the brands need to really get prepared for those moments of search that are coming in. Either they're coming in through Google or Gemini, or they're coming in from, you know, chat GPT, perplexity or Claude.
[00:15:58] And, and that is, this is imperative for brands to, to prepare how they show up on their own website, how they show up outside and, you know, Reddit and social media and, you know, other sort of user generated content that's out there about the brand. And finally, how do they show up in marketplaces? Because these engines are looking for all these sources of information to give that, that consumer the answer that they're looking for.
[00:16:27] And if you don't have your data in order, or if you don't have, you know, people talking about your brand outside of your own website, then you're not going to be discoverable. So that is, discoverability is changing fundamentally. And so brands really do not have much of a choice. And they really do have to invest in that today.
[00:16:49] And I think that's a, that's a great point, Sonal, about the like investing in internally in these changes that are happening with AI. And it's, it's not a nice to have, it's kind of a must have and a must think about. And, and Sonny, I know that AI is affecting a lot of what's happening internally with your teams. And a lot of that is upskilling, right? Like understanding why AI is important. How do you use it?
[00:17:15] So how are you thinking about kind of upskilling and education inside your organization so that people can be ready to use these tools? Yeah, we, we have had so many experiments in a company since two, three years ago when AI appeared. So the first thing initially what we did was really, was to create an AI committee.
[00:17:41] So this AI committee was aiming to, to initiate the projects and to connect with the, the, the business and also the team's skills to, to use AI. And the committee, the committee years ago had members from UI, UX team, product team, IT team, and also marketing team. So like we had a few members who really contributed to that. And we learned so much.
[00:18:10] Last year, this committee transitioned to our AI core team. The reason why we had this transition was really to, how to say, like to, to engage with like more teams than people, because we didn't want to limit like the contributors or the value, the AI drivers to, you know, only a few people. So that is what we did last year.
[00:18:35] And with this, the leadership from this kind of new AI core team, we are really working with a bottom up approach, which like, which means that we want the, the involved, we want to involve and the involvement and the participation from all of our employees regarding how are they using AI to benefit their work? To, you know, like make everything faster, simpler, and more efficient.
[00:19:05] I totally don't believe that I should be the person, you know, organization who knows the best or the most about AI. And, you know, regardless of I believe so or not, I'm not, you know, I'm not this person. Right. So I have learned so much from our teams that when we have such a bottom up approach, like working with AI, because all of the teams,
[00:19:29] and, you know, like the employees, they're like working on their daily work, they have the hands on experience regarding the needs, regarding the reasons and regarding the solutions. And we sometimes now every other week, we meet together every other week, the team with the core team, and we open the meeting to everyone in the company, whoever is available, welcome to join the meeting.
[00:19:54] Sometimes with this biweekly touch base, we review the top identified AI initiatives in order to understand the status, the implementations, and the value connection with the team, with the business. And sometimes we just keep it random. We just want to leave half an hour, for example, to learn from anyone who has, you know, like some interest in sharing. I'll give you an example.
[00:20:22] So a couple of weeks ago, when we like had a kind of like a random meeting to learn from each other about AI. And then one of the teams, it was our product team. So they mentioned that they are using Figma Make a lot recently. I guess you guys know Figma Make. And to visualize the reports, because as I mentioned earlier, AbyDirect is very digitalized.
[00:20:48] Everything we do is heavily about numbers, data, and insights from them. But sometimes people were bothered about, oh, we have so many numbers, about how do we translate the numbers into visuals, into an easy way to let people understand the comments or the takeaways from these numbers. And then, you know, they shared that they recently used a lot of Figma Make. And then they give us the exact cases regarding, oh, what did they do?
[00:21:18] And how did they make it happen? And some examples. And our teams, like some of them were already using Figma Make, but some not. But we really learned a lot from there. And it's an easy tool to use, you know, after you know it, you can do a little bit of research work. And then, you know, like you can start using it pretty, you know, like straightforward. And then now in a team, I heard from them that, you know, like it's already replicating.
[00:21:44] This experience or usage is already replicating to some more people. Yeah, really the philosophy is to, you know, like leverage or like engage our people from like bottom up or top down. We have a combined strategy for that and to use everyone's knowledge and experience to like do the business or the work with AI. And Sunny, are you measuring the usage of AI internally?
[00:22:13] Or are you thinking about incorporating some sort of usage metric or value metric to see and keep people accountable? The answer could be no, but I'm just curious if there's any measurement you're thinking about internally. No, for sure. So as I say, you know, like when I say bottom up, it doesn't mean that we do everything just with this random case or approach.
[00:22:37] Right. We also have a very clear like I would I don't want to call it a top down, but, you know, we have clear goals or objectives when we work with the AI. And then, you know, as I just mentioned earlier, every other week when we meet with the different teams or team leaders, we reveal, right? We reveal all, you know, the status of some business projects when we leverage AI.
[00:23:01] When we talk about this type of business project, we always have clear KPIs or goals regarding the value generation, the implementation timeline, the stakeholders, the owners, etc. And we want to measure the status of progress down the road.
[00:23:20] You know, so often in the media, the first use cases that that people will go to and report on is is efficiency, is, you know, taking air out of it, taking friction out of processes, etc. And I'm wondering, as CEO, when you think about the business value that you want to drive from these AI investments, where does growth come in for you? Like, how much have you focused on the top line impact that AI can?
[00:23:49] And is it are you finding that it is showing some of that potential to help you drive your growth now and in the future versus the the efficiency play that so many people are starting with, it seems like? Yeah, that's a good point, Peter. This is like permanently something we have to balance, right? We wanted to have an immediate understanding of the performance that is contributed by AI.
[00:24:15] But in the meanwhile, we are cautious about the efficiency because it can happen that very often that we invest a lot of efforts and time. And then finally, when we look at the outcome, we it's limited. It's a very, you know, like we have we we we have experienced so many such cases in the in the past. But first of all, to answer your question regarding do we already see value or performance from there? It's absolutely yes. It's no doubt.
[00:24:45] You know, I give you example, another example that like we are right now doing. It is to analyze the customers' sentiments and feedbacks through the NPS, right? So iPad Direct is selling like almost the number. Maybe I don't want to be so specific on, but we are selling millions of pairs of glasses every year.
[00:25:10] Right. And because of that, we are receiving so much comments and feedbacks from customers regarding the products, the quality, the website experience, the messages, right? The brand communications, et cetera. It can be easily overwhelming when we receive so much feedback from customers. And we don't understand what are they really telling us, you know, like if we don't categorize them, process them.
[00:25:39] And like one of the projects this year where actually we started the last year, but this year it's ready to deliver the performance value. It's really, you know, like to automatically upload all these comments or feedback information from our customers and to process them, watch them and organize them.
[00:25:59] And then, you know, finally, we have a unified consolidated report to tell us or teach us that, oh, you know, like we have some pinpoints regarding the experience, maybe starting from the deliveries speed, for example. And then, you know, it's about the post purchase services center. We learn from there.
[00:26:17] And then, you know, like if we want to for some specific area, we want to see a little bit more details regarding what are really the specific points about these issues or concerns from customers. We can easily click one banner or like one tab from there. And then it brings us to the individual comments that we received from customers.
[00:26:41] And so this way that AI is really helping us to make the report, I would say not only integrated, but also deep and detailed enough in order to have some hands-on experience or collection of the messages that receive customers. Then from there, we can look at the actions we should take in order to improve.
[00:27:03] So you can imagine that if, you know, we don't, we didn't have AI, such work can be so manual and so heavy and complicated, almost right, impossible to do. But, you know, like coming back to the point, Peter, that yes, it's a huge potential. And there are still many abilities that we are exploring or like we are ready to explore.
[00:27:27] It's, you know, like, again, we are also like very, how to say, cautious about the efficiency. And we just want to make sure together with our AI core team, we have a management of the balance. I love that.
[00:27:42] And, and the, the, as you mentioned earlier, the culture that you're building, the muscles to be able to, to identify these things, take some advantage of it, test and learn, stop what's not working and double down on what is. I mean, that, that laying that groundwork is going to serve you so well, because this is going to happen and keep happening so fast.
[00:28:05] So having an educated and motivated culture around it, I think is such an inspiration to the industry. And speaking of inspiration, you will be at the LEED conference, May 20th and 21st, speaking on this very topic in New York City. So thank you for this and for also giving back to the broader community at the LEED. And, and Sonal, I wanted to close out on the last question with you.
[00:28:35] The adapting to this, you can hear the investment that EPD has made in thinking thoughtfully about how the organization needs to evolve. How are you seeing those organizational transformations happening across, across the LEED's community? I think the, one of the most essential part about AI transformation is that one, it has to be top down, right?
[00:29:02] It has to be, you know, the leadership has to say, this is how we want to change the organization. And that's so unique about IBI Direct is that, you know, the CEO, it comes down from the CEO all the way down. And that's how you sort of make that culture and, you know, technological change happen. The other thing that, you know, we're hearing and noticing is that a lot of folks have been in this pilot test this, you know, let's test that sort of phase.
[00:29:29] But now they're really, really ready to sort of operationalize it, to really get the value from the investment, you know, and move on to the next phase and change how they go to, how do they do business. And I think that's starting to happen a little bit. And we're seeing a little bit more embrace of AI in, in more deeper processes of the organization. But it's still somewhat following the path of a regular digital transformation.
[00:29:57] But this technology is still so unique and so, you know, unprecedented that it's going to reshape the entire organizations altogether. Right now, it's about how do we get more efficiency out of this process or out of this organization. But in a year or two from now, it's going to be what the agents do versus what people do. And so I think we're going to start to see some of that transformation.
[00:30:23] And we're probably going to have some of those conversations with some of the panelists at the LEED Summit about, you know, how are they sort of making sure that as they're going through the transformation, they protect the brand's essence. Right. A brand that ultimately has to differentiate itself. And so how do you protect the human creativity and essence of the brand as you do that? And the final thing I want to say is there is no map for this. Everybody's sort of trying this out.
[00:30:53] And this is evolving as we speak. No consultant has the answer for you. You have to go to events like the LEED Summit and learn from other people that are in the same boat. Talk to them, hear from them from stage and just network and see where you are and how others are dealing with it. Because there is this is all new territory. And we just, you know, we just have to learn from each other, essentially. Yeah. Sunny, you are nodding. I'd love to hear your double down on that. You agree.
[00:31:23] Yeah. Yeah, I totally agree with that. So like it's such a young or new thing. Right. As we're all learning and exploring our way of working down the road. So I just want to stress the point that, you know, like sometimes we learn a lot of knowledge and information about the AI related things. But really my thinking or advice is to be actionable.
[00:31:52] You know, like sometimes maybe no need that there is no need to overcomplicate or overthinking of the thing. Because as Sona mentioned, and I am totally aligned, that every business has its own unique situation, own unique request, demand, and the expectation from AI. Right. Sometimes some of them are a little bit mature or deeper.
[00:32:18] Some of them may be a little bit, you know, like basic, different starting points. But really regardless of where you, who you are or where are you, just take the actions that you think fit your organization or team. When you learn more, you will reach to the best practices for your business. Yeah, I think, and Lauren, I'd actually like to tag you for this because you've written a lot about organizational transformation in this arena.
[00:32:46] And part of what we're seeing is that it's going to become even more difficult for silos at organizations to stay siloed. Yes. Because the pace at which change is going to happen. And also, you know, I thought of an example earlier. We were talking about the fact that brands now have less control over the consumer shopping experience in an agentic conversation or environment.
[00:33:13] And third party sites are gaining in power because they are the sort of the validation step that AI goes through. And that means brand and direct marketers and e-commerce and data people need to work together super closely for to be able to exert influence across all of those things. And that's not what we often see.
[00:33:38] And Lauren, I was just wondering, when you think about that, you know, what are the organizational, you know, in the last two minutes? How long do you have, Peter? Well, I can send people the webinar, but for the little thing, you know, what is that? What is the big takeaway for why organizations need to shift to respond to that? I think the big piece is that silos aren't optional anymore because everybody else can do it without silos, right?
[00:34:07] Like think about the smaller players, the smaller brands who started yesterday or were digitally native. They don't have silos. They can move faster. And right now you need to be able to move fast in this world to be able to keep up, to understand trends, to be the product that's being recommended by the LLM. So it evens the playing field much more than it has in the past. And I'm actually excited about AI breaking down the silos because it can enable communication cross-functionally, right?
[00:34:35] Like if you have, for example, I see a lot of brands have something like a super agent, which is trained on their specific brand and fed data about every single function. And it can alert someone in sales that there's a marketing campaign happening and maybe we shouldn't run retail media ads on this new product that's being launched. And we wouldn't have known that before because you were siloed and you didn't have a central place or way to communicate that.
[00:35:01] So there's a lot of exciting things happening that AI can help reduce the amount of silos. But to a lot of the points that Sonny was making, this is a huge cultural shift. Every single function is involved in this cultural shift from HR to sales to marketing to supply chain. And if everybody's not on board and thinking differently, it doesn't matter if some of these changes are put in place.
[00:35:27] So Sonal, just to close out, I want to make sure people who are listening can find their way to check out the agenda at the LEED conference. Where should they go? Yeah. So the conference is on May 20th and 21st in New York City. The agenda is now live and it's going to be at summit.the-leed.co. Check it out. We currently have over 70 speakers and adding more every day.
[00:35:57] And so, and a lot to do with AI, both from, you know, how to use it well, but also how to organize around it and how to inspire. A lot of sessions on that as well and how to protect your own career in this age of AI displacement. So a lot of conversations on that and other things. And again, Soni, thank you so much for being generous enough to come here to share this with our listening community. In addition to the contribution you're going to make at the LEED.
[00:36:26] Thank you so much to both of you for joining us. We're really grateful. Thank you for having us. Thank you both so much. Thanks to Soni and Sonal for their useful insights. Thanks for being part of our community. Thank you.