So for the last part here, I have a demo that I actually recorded yesterday just for the sake of having it perfectly done. You'll see there's a date on the twenty fourth if you look carefully. So the demo will showcase all the brands of Caleras, but not all of them, but, some of them that I handpicked for, specific features. We'll also go after that in Coveo cloud. You see on the back end, it's gonna get quite a bit technical, but we're gonna see the new features that we have developed in reaction to everything that that happened in the Hilarious project. So we talk about the complexity of the catalog. We're gonna showcase how now we have a full system in Coveo call the commerce catalog to handle these relationships, but also how to do better merchandising with promotion, AB testing, and even reporting at the end. So every tool that we have developed for Caleras, they're not available for, all clients, and then we're gonna showcase that. So I'll leave you to this demo, and then we're gonna come back for a quick q and a session just right after. I wanna showcase now Coveo for Commerce with live examples of the Calera's brands. Let's get started here with Franco Sardo. I will use the mega menu to browse one of the listing pages. They're all powered by Coveo. And once you reach a listing page, you can see that the simple yet elegant interface is packed with relevance and features. The user can use filter to select sizes on the variant level. They can also filter on attributes of the product level, such as color, or also other attributes at the variant level, such as width. Once all filtered, you'll notice that the products are grouped together. This is to save real estate as you can browse for many different variations of the same product whether it's a different price or a different color. You can see that we're also bringing back some content that is not product. So these promotions here are a nice shortcut to reach other part of the website and makes the experience more enjoyable. Another cool experience is Ricca. This brand has a completely different look and feel. It's a sports brand, and the architecture behind the scene is the same. So if we use the mega menu, we're gonna reach a listing page powered by Coveo. There's less product attributes here, so there's less filter, but the same functionality we saw such as filtering on variant level or on the product level or available. You see here a really good example on how customizable and scalable is Coveo UI in Coveo for Commerce. We'll now head north in Naturalizer, Canada. Kilaris has multiple brands, but it's also operating in different countries such as Canada with different languages like French. Folks here in Quebec City at Coveo were really happy to finally be able to search for winter boots in French Canadian. With more than four million visitor every month, Famous Footwear is the most skilled experience at Calaris. Coveo is everywhere on this website. If you scroll a little bit down directly on the homepage, you'll find some recommendation widgets. They're using here top seller, but we also have available top brands or top viewed product, and these components can be drag and drop across the interface by content authors directly. The same architecture of the other website is available here. So if you use the mega menu, you're gonna reach listing page. But then I want to use the query suggest here. These are powered by previous successful events on the website, entirely user generated. And if you click on one of them, you're gonna reach the result page. This is where we get the highest level of complexity. Like all other brands, you can select variations such as sizes and width. But here we also have gender as a new, factor where you have men's shoes, boys, women, or girls shoes. Let's select men. The best feature out there, in my opinion, is that you can select a specific size, and you see how fast it's reacting. And based on the size and the gender and all the other filters, you can go down and select the store you want to pick it up from. At this point, you're sure that you don't have a bad surprise when you reach the PDP, as everything here in terms of sizes and models are available in the UI. On the product page level, quite standard here for ecommerce experience, you have the variation of color and sizes. And then if you scroll a little bit down, you'll see all the recommended products. This is powered by machine learning, and it helps actually to automate the content authoring experience as folks at Caleras don't need to manually cure these recommendations. They're really based on purchases, clicks, views, and all the other signals that we're receiving from the interface directly. Enough showcasing. Let's now go in Coveo Cloud to see how we're building these experiences. The backbone behind the complex query system is called the commerce catalog. This is where you define product variance in stores and manage the interaction between these different objects. Let's go and create a catalog from scratch. We'll call it r three sixty. A catalog is, junction between configuration and the products themselves. So first, we're gonna select the products. They are in a very push source called our three sixty shoe company. Then for this specific catalog, you can set up what would be included. Does it have variance on the product level directly, or do you have availability such as stores or even customer configuration can be reusable across different brands. In this case, The configuration can be reusable across different brands. In this case, I'll call it products and variants as I think other, brands configuration later down in my deployment. The next step is to map what is a product and what is a variant. In my case, I push them using the object type product in variant. So the mapping is extremely simple. You select your product. You select what is the primary key of this product. And then on the variant level, you do the same actually. What is a variant? And you can select the primary key for your variant. At this point, the user interface will display product and underneath them, all the variants that are available. So you can use the search interface here to validate that everything is okay, and then we can go on the next step. The last step here is to map to standard fields. What we're doing here is taking the metadata from the JSON that you sent, and we're mapping them to generic fields. This is really good for machine learning purposes, but also for a bunch of other features that will be shared across all clients as we're making everything uniform and standardized across all deployments. There we go. Our first catalog is created. A lot of new feature that we have to offer today are located in the query pipeline. If we select the query pipeline that matches the use case I'm building, r three sixty, you'll see here that a little rehaul of the interface happened. We have new feature such as groups and campaign. Groups and campaigns are the best way to to organize and make everything clean in your pipeline. You see I already have a permanent group here called base relevance rule. This is used to group together everything that will be applied to all queries, such here such as here a boost on the sales item and demoting low inventory item. Let me show you how to create a campaign for a promotion. Let's create a new group here and call it Easter promotion. I will set a date, a start date, and an end date. So let's go until April thirty here, and then I will save the group. Once a group created, if you go on result ranking, you can create a rule and assign it to that group. So I will create a feature result in this case. The featured result will take one of the product and push it at the top automatically. Let's call this one blue shoe as this is what I wanna promote. I will select eastern promotion as my groups, as my campaign, and then I will go here and select an item in my index. Let's take this blue Lacoste sneaker here. You can use the small search interface to test that your rule is working accordingly. And if you press save and the group is active, this will be pushed in production instantly. As a part of these new features for commerce and the pipeline redesign, we decided to include AB tests directly at the pipeline level. Let's go here and showcase how this works. If you click on AB test, you can configure an AB test on your pipeline. The only thing you need to do is to set the percentage of traffic split between the original pipeline and the test scenario we're building. For building the test scenario, we can just use here the edit button. We're gonna have access to all the features in the pipeline, and we can add one that we want to test to make sure that there is an improvement in terms of relevance by deploying something new. Here, I want to add a new machine learning model. My team now is tracking more events on the front end, and I want to use these events to make, automatic relevance tuning better. So I'll select my ART model associated to the pipeline, make sure that it's in the first position so it's tested thoroughly, and close this panel. With that being said, my AB test configuration is ready. The only thing I need to do is confirm and press play. All the metrics will be available directly in this panel. And once you're done with your e b test, you can stop it and decide to discard the test scenario or reuse it and deploy a new feature in your pipeline. At this point, if you go back in machine learning, you'll see my menu model is now the first one in line. The last feature I wanna share today in this feature blitz is the commerce dashboards. This is an entirely new way to report on your commerce implementation, and it's entirely commerce focused. You can see here the number of session, transaction, unit sold, and revenue, also conversion rate, average order size, average revenue per unit, and average order. An interesting part is also the orange here, which is the percentage of, influence by Coveo regarding all these metrics. Really good way to do attribution. The commerce dashboard will also let you browse your conversion funnel so you can see how many sessions with product clicks, with add to cart, and then at the end with transaction. Really standard funnel here. Each query can be explored. So you can see what's the percentage of total search regarding shoe or Nike here. You can see what are the suggestions, click rate, total revenue, and even revenue per search. The catalog performance is another way to display what's happening, and you can see the top brands and their influence over what's happening on the commerce deployment. Last but not least, there are some timelines, so you can see your revenue spread through time and also the conversion funnel for all brand. Thank you very much for listening this demo, and see you later, folks.