Hi, everyone. And welcome to this session where we will talk about the impact of AI in e commerce. My name is Sheila Moray. I'm the chief marketing officer at Coveo. It's an enterprise AI platform that enables the connected, trusted personalized experiences at scale with AI search, AI recommendations, one on one personalizations, and generated answering. And today, I'm thrilled to be joined by Venda Provato. She's the esteemed chief marketing and digital officer at LCBO. LCBO stand for liquor control board of Ontario in Canada. And one of the largest retailer of liquor in the world. So we will learn a lot about it today. But you also you're known as a trailblazer of brand transformation you work at second cup, coffee, while you did the full transformation of their brand. You also work at indigo book books and AOL Canada, an amazing carrier. Well, let's start about let's start to talk about LCBO. Thank you, Sheila, and thanks for the opportunity to be here today. Yeah. Tell us a little bit about LCBO, the business you're in that what the typical customer look like. And, how do you engage and reach them? So I'll give a little bit of context for those in the room that maybe are not as familiar with the LCBO. We've actually been around for a long time, ninety seven years. And as Sheila said, we are one of the largest wholesalers and retailers of beverage alcohol in the world. Our annual sales are more than seven billion dollars. From a b to c perspective, we operate close to seven hundred lead, retail locations in the province of Ontario. We also have about four hundred, licensed convenient outlets. We carry over thirty four thousand products from around the world. And then, you know, from just a B2C perspective, again, we, you know, getting to an LCBO store or one of our outlets is really just a very short drive or a short walk for most of our customers. You know, you can get to an LCBO in maybe five or fifteen minutes. But the digital presence is a really important part of our business as well. We have websites, we have mobile apps, most of our customers actually start their journey with us online. So even though sales is a very small percentage to percent of our retail sales. It is where a lot of customers come to start their journey of shopping with us, to browse, to look up products, WE DO SEE ABOUT A MILLION VISITS TO OUR WEBSITES EVERY SINGLE WEEK. AND IF YOU CONSIDER THAT, YOU KNOW, THERE'S MAYBE ten or twelve MILLION drinking HAD population in Ontario, that's actually a really big number. And so for that reason, our digital experience is a really important part of the business. We launched our new website in twenty twenty two, and that was with Adobe and with Coveo. And, we're actually in the process of launching a B2B experience on our website as well. So that's a little bit about the business. Amazing. And you all know it in the room. Customer experience is really today competitive frontline. It's a make or break for business. And what we'll talk about today is the fact that it's not just online or offline It's omnichannel. It's all connected. And I know you've you're feeling that and you're seeing that as well. But what other behaviors have you seen that were at that influence that digital transformation at LCV. Yeah. I think similar to a lot of other retail organizations, a lot of our customers shop with us both in store and online. And like other retail organizations, an omnichannel customer is you know, more valuable, spends more money in our case almost two times more when they shop with us in both channels. Our customers that shop us in store, about two thirds of them told us that they also visit us online. And so we know they're seeing there's coming to us in both channels. And so our e commerce vision is really about making sure that we're offering a personalized experience an easy, integrated, lots of different options so that the transition between physical and digital is seamless. It's also very much tied to our overall mission and vision and very closely linked to what our brand promises. Which is all about providing, you know, perfect choices made really easy for our customers so that they can have a great experience. This is gonna become even more important for us in the coming years with the announcement recently from the Ontario provincial government really open up the points of retail distribution. So in the future, in the next couple of years, customers are going to have a lot more options, thousands more options of where they can buy beverage alcohol, in convenience stores, in big box, retail locations, spending grocery, so making sure that we're providing a really great experience connecting people with what they're looking for This is critical for us to keep, you know, people interested to shop with LCBO. Yeah. And you know that, you all know that customers have high expectations for experience. More and more, they're looking for experience that are personalized relevant to them. They don't wanna waste their time online. They want to look get the product they are looking for fast. And they're expecting brands to be proactive, prescriptive, to be there for them. So they are actually getting the product that is that they needed in the moment. But I know that going through those digital transformation has a lot of challenge. I'm sure you have you faced a lot of them. So what are the challenge that stand out the most for you in this transformation? Yeah. Definitely. I think, you know, with our old e commerce platform, we didn't have the benefit of AI. There wasn't any machine learning. So all the search functionality was quite basic. And it was really being directly managed by us. There wasn't continuous learning, there wasn't improvement. So as you can imagine, customers were spending a lot of time searching for what they were looking for. They were doing a lot of scrolling. I think the other disadvantage was that they had to be very precise in what they were putting in as a search term. So the proper word spelled correctly. Otherwise, they wouldn't be served up the right result. And so as you can imagine, we saw that reflected, that dissatisfaction, re reflected in our E NPS, in our NPS scores and also in, in customer satisfaction. And we have so many different products, to scroll through. It's not a great experience. And and we didn't have the opportunity to do anything in the way of personalization either. And so I think, you know, the other challenge we had, w, like many other organizations that have been around for a hundred years, there's a lot of legacy systems that we have. So systems that maybe don't enable us to do what we need to do. They aren't fully integrated. And so while we did a really great job, I think of updating the front end of that online experience, we still had challenges where we weren't fully able to realize the benefit of that investment because for example, we didn't have, real time inventory and so, you know, the good news I think is that we have embarked on a future state modernization, journey, And so what that means is that our back end systems, our business processes, all those things are gonna get modernized. And so in doing so, I think it'll really enable us to deliver on our, on our digital vision. I also think it's really great that we are doing them concurrently. We're not waiting to update digital experience. We're doing it concurrently with updating our back end and our business processes because we're already starting to see the benefit of some of those changes we're making, and I know we're gonna talk about results later. Yeah. And today's session is about AI. You came to hear about AI and commerce. So let's dive into this. But before, I mentioned that our CEO at Coveo always say that we should not be afraid of losing our job because of AI. We should be afraid of losing our business because our competitor is going with AI. So that's a different mindset. We should not be afraid of technology and AI. We should embrace it. And there's this AI experience advantage you're getting if you're able to increase customer satisfaction with your experience while increasing your business outcome, while increasing your revenue, your revenue per visit, your margin, your conversion, you can do both with AI. At Cabello, this is a a list of AI models we have part of it. And it can go from educating your customer online, giving them the content they need to product discovery, to, of course, recommending the right product in the in the moment up to customer service and post sales. So what are the models you actually launched at LCBO and how did it benefit it the business. Yeah. I think, you know, first, I'll say that, personalization is a journey, and I would say, you know, AI, of course, is an important enabler for personalization, but AI itself is also a journey. And so for us, I mean, we're at the beginning of our journey, but it really starts with, as Sheila said, creating relevancy. Because at the end of the day, what we wanna do is connect people with things that are of interest to them. And that directly ties back to our brand promise that I talked about earlier. And so, you know, the people are really important in that work because what they're doing is they're leveraging the data, they're leveraging the insight to create those more personalized experiences. Yes, we're leveraging the machine learning, but it's not just on the machine, we're making decisions around where and how to improve the experiences. I think that the other thing we recognize is that while people do value generalization, they are also, you know, very protective of their data, and they, you know, they expect that, you know, their privacy is being did as well. And so they're not always willing and forthcoming to share information. And they aren't always most of the time, they're not even logged in assess when they're on our website. They'll come and browse, they'll transact, they'll check out as a guest. And so even though we know a lot about our customers who have an online profile who are maybe in the loyalty program, which is aeroplan for us, unless you are signed in, and we know who you are, it's hard for us to create that personalized experience. And I think this is where AI insert and understanding the intent of what people are doing on the website and what they're looking for, being able to leverage that data and serve up content, serve up products that are relevant while they're in the session is equally as important. One hundred percent. You don't you don't need to be authenticated to be to get the personalized experience. In a few clicks with Coveo, you can get a personalized experience without having the full information. Yeah. And I I think just to touch on, because I know on the slide behind us, there's there's quite a few of those different points along the journey. I can speak just a little bit maybe to where we are today. So if we look at semantic search, I think this is a good one because this is really about you know, creating contextual clues, if you will, to really try and determine what the intent is of the search that the person conducting. And I think a good example of this is, we have sorry. Go back to the previous slide. We have a lot of, we have a lot of, products that we sell that are celebrity brands or celebrity, backed products that there's a lot of interest around. You guys may be familiar with Ryan Reynolds, very popular, Canadian actor, and you may or may not know that he actually has a gin product. You may or may not know that Jen is aviation Gen. So sometimes people will come to the website. They won't remember what his product is, and they'll start typing in Ryan Reynolds. And so by creating this contextual connection, we actually are able to serve up aviation gin, as as the product that is most relevant and probably what they're looking for. In query suggestions, this is really a search experience that, displays a list of maybe possible, recommendations or possible suggestions to whatever the search query is. And while that person is typing, maybe they haven't even finished their thought yet, but because of all the learning that we have around prior searches, it's already to populate a list of options that customers can choose from. Automatic, relevancy tuning or ART. This this is, I think critically important because what it's doing is it's not only analyzing what people are searching for, but it's actually tying it back to an action that was taken as a result of that. So did they click through on a product from the search page? Did they put it in their cart? Did they buy that product all that information gets fed back in. And so the next time a product search is done, it's just improving the relevancy of the products that are being served. And then finally, content and product recommendations, again, looking at actions that were taken prior, how do we serve up product and content that we know is going to deliver a desirable outcome. All of this is really important because again, going back to people want to be able to find really quickly what they're what they're looking for. They don't wanna spend time. No. Exactly. And they are looking for experience that are in the survey we we did, people are looking for online experience that are that are as good or better than the offline experience, the in store experience. But for LCBO, in store, you have your wine expert, your liquor expert, that can help them, but you don't have that online. So I know you're moving to a much better experience. You are in your transformation. So where are you today in offering this amazing experience that is as good as in store? Yeah. I mean, you're right. It oh, just got really loud. I think you're right in that, in store, a customer has the benefit of talking to a very knowledgeable sales representative or a product saltons, they can immediately take you to the product that you're looking for. They can make great recommendations because either they know you, because you're a regular customer, or they can ask you a series of questions that helps them to make a better recommendation. This is really the experience that we're trying to mimic online. I think, you know, using AI, we're able to take that data, take that learning from prior searches and prior actions, and really serve up that recommendation or that relevant, that relevant search result. I think if you click and play the video here, I think folks can just see AI in action on our site. So a person starts typing in the search box, the suggestions immediately start to appear, even if, you know, they don't have the right spelling. And as they the typing progresses, those suggestions actually get updated. They're the relevant. They're the most desirable outcome recommendations. They can even, refine sorry. They can even see recommendations in products, in recipes, and in content as well. We've got really great content on our site. A lot of it comes from our food and drink magazine, which we publish five times a year. And, that's integrated into searches as well. The customer also has the opportunity to refine the search so they can look at country of origin. They can look at price point. They can look at sugar content. All this stuff. We know really important in the decision making process. So and they can also actually click on and say just show me, results for products that you actually have available that I can buy, either in line or online or in your stores. So, again, I think all this all this content, all these product recommendations they get really smarter because they're leveraging the searches and the actions that came before. Amazing. That was a great journey you've been in. So many in the audience today are at different point in their digital transformation. Some are still doing a lot of manual rules. Some have AI, but they don't really know. They didn't turn it on. They're they wanna overrule what AI is doing. So What are what did you learn to that journey? And what can you what advice can you give to the audience? I think, you know, our biggest learning, you're right, is in that just let the model do its work. I think initially, we had programmed in a lot of redirect to products and to pages and to content that we wanted the user to see as opposed to reflect to what the user was looking for. The other downside of that is when you add new products to your catalog, they may not automatically show up on their own. So I would say that's the biggest asset. Paying attention to the language as well. We really continuously monitor, how people are searching how people are spelling things so that we can improve the quality of our search results. It's the search is also a great place to get insights on what people are looking for sometimes. Those are products that we don't carry. So we use that information in a couple of ways. We'll share it with our organizing team so they can determine whether or not it's a product we should be carrying. And even if it isn't, we can use that to say what other products do we carry that are similar to that that we can actually serve up in the results so they don't come back with, you know, null searches. And then the last part I wanted to touch on also equally important is, you know, finding the right place for featured results. We give our suppliers or trade partners an opportunity to pay for placement in search. Sometimes it's tied to promotional campaigns that we're doing. A really important part of driving marketing revenue. But I think the learning there is that you don't wanna serve up those results, a head of whatever the most relevant result is based on the intent of what the person was looking for. Great learning, great advice. But now let's talk about results. Is it working? What are the key PIs you were tracking? What are the KPIs that keep you up at night and where were you able to see some improvement? Yeah. I would say, you know, maybe feels like we're still at the beginning of our journey. There's lots to do. Maybe these small changes we made are not making a difference, but they actually are making a difference. And I think that's probably another us and learn is just get started, just do start doing something. I think we also think of our KPIs more holistically. It's not just about driving conversion because remember our store network, it's very convenient. People can easily come in and buy what they're looking for. So for us, it's broader than just looking at conversion. However, I can tell you that conversion when there is a search action involved is almost twice as high as a visit where there is no search involved. And that relationship, that ratio, has actually improved with our new site with Coveo versus our old sites. So I think that's a really important KPI. Click through. So from the results search page, so a fifty eight percent improvement, which means person found something relevant, found what they were looking for, and they clicked through to a product page. We also saw a really notable improvement in the ranking So what this means is the results page that served up the results from the search, what the person was looking for showed up much higher on the page, So that meant less time, less scrolling. And on average, that product position has improved by sixteen positions. I think that's really impressive. And then equally important is the customer experience. And what we're seeing in our customer surveys is a notable improvement, a significant improvement in comments related to search and how easy it is to find things on our site now. Right? Well, Now what's next for LTVO? I know you're really ambitious. You have a lot of projects. So maybe we can focus on the AI and ML, but do. We have a lot of things on the gills. So maybe I'll I'll I'll start with, you know, product discovery experiences, and I think this is really where search and I've, AI fit in. So obviously continuous optimization, doing a lot of testing and learning, but the area we're next to really focus on and is enhancing the recommendation model. So, obviously, for cross sell and up sell opportunities, How can we I mean, we do some, recommendations today on our website, but they tend to be more served up as the general products that are most popular, and we wanna refine that. We want those product recommendations and that content recommendation, to get more specific, more relevant throughout the entire session. So what they're seeing when they're on the homepage versus the product page versus in their cart, those recommendations get, more personalized and and just more more relevant. Yeah. And then looking at personalization more broadly, more generally, it's How do we leverage the data as I was talking about earlier? The data we know about our customers and the loyalty program the data we have about them, the transactional behavior we have about them when they're shopping with us online. How do we look at that and really combine it? With the intelligence that we're getting from AI, from what is happening in the session. So how do we bring those two things together? To really create that personalized experience. And then finally, I would say, more broadly, as I talked about earlier, it's really about modernizing everything that we're doing. And for us right now, it is the first step that we are working on in the e commerce team is about migrating all of our digital channels onto one platform. Yeah. So we have a vintage to shop online, which is a separate website from cbo dot com, that we're moving on to the lcbo dot com platform, and we're also launching a B2B experience on that platform. And all of that is really a great precursor. I think to what's happening with the future state modernization, the improvement on the back end systems where Now I think we'll really be able to fully, capitalize not only on the investment, but to fulfill that vision. Amazing. Thank you so much for all those great insights. Last slide is the three key takeaways of this session. Thank you, Ben. That we learn, we learn a lot working with you guys. Thank you for your business. So three three takeaways of this. The first one is search, AI search is central and strategic in customer experience and online customer experience. That window on your website is your window to communicate with your customer. They can tell you a lot in that search box. So it goes from being a search box to an intent box, understanding their intent. It's now going to be a conversational box with generative answering. You'll be able to have that conversation with your customer. It's a customer insights box when you learn from them. So chief search can be really expensive. You need to invest in search. It's important in invest in AI search so that you can personalize as much as possible even if they're not authenticated, there's a lot you can do. Then if you want to offer that best in class personalized experience, there's no way around it. You need AI. You need an AI platform that can help you get there to connect all your touch point, your omni channel touch point together and offer that friction less personalized experience. That means AI search, AI recommendation, personalization of your old experience, and of course, generative answering. And last but not least, AI do deliver results. And you've seen it, and we have tons of other data point that proves that you should go right in. Don't wait. Like you mentioned, start small and scale, but start somewhere because that's the future. You're not gonna be able to win on both side, increasing customer satisfaction and deliver business outcome without AI. So And this is like oh, what happened? I just wanted to be able to say, you know, Coveo has been a really great partner for us as as well to really understand where the opportunities are. And how to get great results by leveraging AI. So thank you for that. Thank you so much. That's amazing. Thank you. Thank you, everyone. We're around if you have questions.