Hi, everyone. Thank you so much for joining our spring release of new and Coveo website edition. My name is Carrie Anne Beach. I'm a senior product marketing manager at Coveo for the website line of business. I am joined by my colleague, Michelle. Michelle, you wanna introduce yourself? Hi, everyone. I'm Michelle, senior product manager at Coveo, in charge of the platform parts. Awesome. So we're really excited to share some of our new innovations with you. Before we get started, just a quick disclaimer. You will see some forward looking innovations, so please just make sure, to make buying decisions based on publicly in, publicly available information. So with that, let's jump right in. And before we share what's new, I just wanna take a moment, to talk to you a little bit about our vision for Coveo for websites. So when you think of websites, the first thing that likely comes to mind is site search. This makes sense. It's a foundational piece of your web strategy. You need to help your customers quickly and easily find what they need. This is actually a really great opportunity for you to convert high intent inbound leads that are coming to your site by delivering relevant content during their moments of need. But, really, this is just the start, and we wanna help you master and ultimately go beyond search. So you can help your users discover new and relevant content with the help of proactive recommendations and even start implementing cutting edge technology like generative answering to help your customers self-service that much faster while enhancing your team's efficiency. And, ultimately, we wanna help you drive an end to end digital customer experience across all of your sites because your website is a lot more than just your dot com. It's really every site that your users are interacting with. Right? From your dot com to support, maybe even a commerce site, Customers and prospects today expect a cohesive and personalized experience at every point of interaction. And luckily enough, you can give this to them using a single AI search and generative experience platform. So on the top slide, on the top of this slide here, you're gonna see different use cases that Coveo supports to create this seamless customer and even employee experience because your employees are customers too. And then on the bottom, you'll see the various sites that your users and stake holders are engaging with throughout their journey. So you may not own every piece of this, but you can bring global site capabilities like content discovery and create that cohesive digital experience. So we're really excited to share some new product updates that'll help you drive that digital customer experience and even future proof your technology stack. And they can be really summarized into two themes of innovation. The first one is about empowering your teams, both marketers and also technical teams, to be more self sufficient to decrease time to value. And then the second is all about enhancing the customer experience by helping them self-service. So we have some really, exciting enhancements around generative answering. So with that, I will pass it off to Michelle to walk you through, some of these innovations. So you can see here the various new features on Coveo that we have either already been released or are being finalized and will be released in the following weeks and months. We won't have time to cover them all, and we'll focus instead on the one that are the most relevant for website use case, starting with connectivity and integrations. So one major differentiator of Coveo is its unique ability to aggregate content from many different sources and bring them all in one place behind a unified index. And to achieve this, you can rely on a wide array of native and generic connectors that allow you to index any software that are relevant for your website search experience. Today, we will focus on two connectors that are widely used for website use cases, web and site map connectors. Both are generic connectors that allow to index any websites. Let me now deep dive a bit into what each of them does and the improvement we just added to. Starting with web. So the web connector is the easiest to set up. You will just give it a set of starting URL, and it will start crawling your website. Each time a page reference another page of the same domain, it will crawl this additional page and expand like this until your full website have been indexed. Now a common issue we have when indexing website is how to deal with the repetitive parts, like the header, the footers, the sign up, the ads that are present on every page. If we take a concrete example, let's assume that you have the keyword documentation in the header of your websites. It mean this keyword will appear on each and every page. And if the user is typing a request with documentation in it, we might return many pages that have nothing to do with documentation because it is located in the header. So fortunately our ML model, they as they get more data points, we'll be quickly able to surface the most relevant result first. But still, you want to implement everything possible to have the most relevant results. So here called the concept of web scraping configuration that exists at Coveo for a long time, and it allows you to define which portion of the webpage will be excluded at the same time. So for example, you'll be able to exclude the header, the footer, and so on so that you index only what is irrelevant for the end users. So like I said, this feature was already existing, but it required a bit of time to configure properly. And we just made that easier with default web scrapping. So here's a short video of how it works. Here you have an organization with a web source that does not have web scraping yet. If you open the source, you will know see a small recommended label on the on on the source configuration. And then if you click on it, you'll basically have a little bit of guidance about what web scraping is about. And if you want to implement it, you will see that it will be pretty straightforward. By activating the default web scrapping, it will open a a model showing you all the CSS, selectors that we exclude, which is default configuration. It means that you can, of course, modify some of them in case, like, your website is as really some specificities that you want to take into account. Otherwise, and most of the time, actually, it should be good like this. You're just able to save, and that's it. Now you're indexing only what is relevant, on your website. So you can like I said, you can always fine tune the default configuration based on your site specificities, and you can also use our Chrome extension to, to to go even quicker on that part. But at least you get a solid starting point that you should highly reduce the time needed to implement. So that was it for our web connector. Now on to our site map connector. The site map is another generic connector, but this one will not proceed like the web connector calling iteratively each page of your website. Instead, it will rely on the site my set map file that references the architectures of all the pages of your sites, and it will crop precisely these pages. And as a consequence, not only the time to index and refresh content will be faster, but it can also, like, leverage incremental refresh to pick up faster changes in few pages only. So for this sitemap connector, not only did we improve it by adding the same default transcribing feature that you saw previously, but also we took this opportunity to improve the overall user experience and provide the same easiness to configure than for web sources. Previously, the configuration relied heavily on raw JSON configuration, which could be a little bit tedious and sometimes error prone. So we improve the source by adding all the most used configuration options directly in the user interface, making it configuration much easier and faster. So in short, for both web and site map, we are empowering technical teams to be more efficient during the configuration phase. Now let's see a couple of additional features about the search experience targeting primarily marketing teams and empowering them to, like, deliver results faster. Before jumping into the specific feature, let's start by introducing the concept of a builder. Since two thousand twenty two, we invested, lots on CoveoBuilder. The whole philosophy behind it is to empower these marketing teams and these business teams more generally to be able to build faster proof of concepts. Whenever you're working and building a search experience, there are often a couple of iteration needed to redefine fine tune until you get the the perfect, end user experience. And this phase could be fast and could obviously take longer if you rely on technical teams to implement all the changes. So that's precisely why the builder are here. Here, we're giving all the tools to marketing teams to do this iteration cycle on their own, and you will see the exact feature that that we we, had developed on that part. Two key principles to keep in mind with builders. The first one is best practices. These builders, they just work out of the box. Everything required to make your integration a success is preconfigured. Machine learning, analytics, and UX is designed as per Coveo best practices. And the second part is that it's innovation ready. The new features are readily available directly into the builder. So, basically, by using Coveo builder, you subscribe to Coveo ability to innovate. Any new features will be easy to add, to try, and deploy. So now let's jump in into two concrete builder example. The first one is a hosted search page builder. So this will allow marketing teams to design, configure, and deploy new search pages from the QBO platform with just a few point and click. This hosted search page existed before, but we just released a couple of, nice improvement to it. So for example, we now have more styling options when configuring the page. You also have the option the ability to activate dynamic navigation. What it means is that the the facet that you see on the left of the page and their value will be reordered based on user query. So that if we detect that some facets are more used than other and possibly some values of these facets are more used than other, they will be surfaced at the top so that, like, user won't have to to to search to scroll and and search them. Thanks to our ML model. And another improvement that we did is we also made smart snippets and RGA, directly activable at a click on a button inside the configuration of this hosted search page. So that's it for the first builder. And the second builder is the in product experience or IPX, which is, used on website for, like, generally support help center documentation. It is a lightweight search interface that can be embedded on any website. In this example, you can see, like, a a small help button that will open a window allowing user to search, for example, for frequently asked questions article from anywhere on your website. So similarly as before, we added a couple of nice improvement to this IPX experience. We added version history, which mean that assuming that you are multiple people working on this, IPX experience, in case there is an error with, like, the the current configuration of IPX, it just takes one click to revert it to the previous working configuration and give you enough time to understood what was wrong and and and then deploy the next working version. Another improvement that we did is we are now creating automatically pipelines and ML model. So that's when you're creating your IPX, you don't even have to, create your ML model, your pipelines, and and link it to the IPX. This will be done automatically and under the hood for even easier and faster implementation. And finally, we also activated RGA. I'll I'll deep dive into RGA later, but we activated RGA just at a click on a button on the IPX. So it it it it's also, like, much faster, to implement. So like I mentioned, post builders, they are really here to help the marketing teams experiment, test, learn on their own in the fastest possible way. But, generally, what happens is after the cycle of experimentation, you might probably want to go for a more custom implementation because, your site had some specificities. Maybe, you want also to keep its styling. So in this case, you might turn to, our different libraries that we offer. And these library are really designed to, make implementation easier and quicker. As you show on this graph, you have three options with different level of simplicity and customization. The first level is using Coveo atomic. So it's the easiest and fastest to implement with a set of UI components for assembling responsive, accessible, and future proof Coveo powered search interfaces. Now let's say that you need a bit more customization. For example, you want to be able to use your own UI. In this case, you might go for Coveo headless, which is a bit more complex to implement, but it's still a major time saver as it acts as a helpful wrapper around Coveo APIs. And of course, there is always a possibility to use COVID APIs directly for maximum flexibility, but obviously at a cost of a heavier implementation. So now that the difference between the three is clear, the the main news and improvement that we are bringing to these libraries is that we are deprecating the v one of our CLI headless anatomy, and we really soon the v of these libraries. And among other things, this v3 will facilitate the implementation of the new event tracking protocol. More on this later, organization endpoints, server side rendering, support in React, and more. So we have seen the various improvement to save time for marketing and technical teams who are creating search experiences. Now let's have a look at another set of improvements designed to increase end user self-service and satisfaction. A few words about RGA first. So in case you haven't heard about it, Coveo has released in, in GA end of last year a new relevance generative answering experience, that can be available across all channels to enter a consistent and relevant end to end user experience. So RGA allows using LLM to provide answers to complex questions in human language. We are removing hallucination risk by making sure that the answer is based only on the relevant content inside your unified Coveo index. And it also means that it this is able to synthesize and answers across a large number of sources. And finally, in case a user want to dig further, we will also reference these sources. So as said before, RGA can be deployed on any search experience to follow your users across channels, but we continue improving it over time. And I will show you a couple of these improvements in the following slides. The first one is follow-up questions. With follow-up questions, you will be able to elevate your user self-service game by providing instant gratification to your users. Why should we settle for unanswered queries where you can delight your user with instant solutions? Our new feature boosts the answers rates and deflect cases and help users self serve while improving their satisfaction. So what are follow-up questions? Well, as you can see on the screenshot, they they they are, small reformulation of the initial questions that will help user find deeper answers and ensure that they find what they really need. And how does it work? It is powered by our LLM technologies. So the follow-up question, they are intelligent generated violence of the initial query, always leading users to the right solution. The other first improvement, we're also working on conversational experiences. We'll take the user self-service and satisfaction one step further than the follow-up questions. Not only in we it will suggest follow-up questions, but the user can also type its own follow-up questions. And in both cases, it will keep the context of the previous answer in the subsequent queries. It also leverages our behavioral analytics to retrieve the most relevant documents and generate precise answer. So as you can see here, we really have on this short video, a conversational experience when starting with a question. And then as Coveo answer, the user is able to ask precision along the way. And like I said before, Coveo will always keep in mind the answers to the previous questions, for the next ones. Onto rich text formatting now. What happens sometimes if the the user asks for complex question is that you might get an answer that that, like, includes a lot of text. And we wanted to make this content more digestible from the end user so that we see more engagements and, and and basically to have them find their answer quicker. And that's precisely what we did with rich text formatting. Instead of just, displaying plain text, we'll not be able to format it. For example, adding some code blocks, adding some tables, adding some bullet points, some bold, some italic, again, everything that make the, the, the answer more digestible, for the end user. And thus, again, improve engagement and self-service. And finally, not but last but not least, a lot of our customers have different sites with different languages. And of course, we wanted to be able to deploy RGA for these other languages as well. And that's what we are working on at the moment was the ability to, use RGA in French, Spanish, Italian, and German. So all these feature will greatly enhance user self-service and satisfaction while improving efficiency for your teams. Let's now look at a couple of improvement to streamline implementation for technical teams while ensuring better data quality and reporting. The new event tracking protocol is designed to feed our ML model with highly qualitative data. This, of course, will result in even more relevant results for our end users. So this new event tracking protocol is designed a new event schema that are carefully thought for Coveo requirements and that don't accept custom data. But we also wanted to make sure that these better data quality does not come at the expense of implementation time. Actually, we wanted to do exactly the opposite. We wanted better data quality that takes less time to implement. And so to do so, we will also release a new client library relay, a Chrome extension that will allow you to validate the event syntax validation as you implement it. And finally, we will also ensure that automatic search events are created directly by our search API without requiring implementers to create them. And this is a small zoom on the, Chrome extension that that that is available, and that I mentioned previously, that you can see in action, like whenever, triggering an event in the UI, you can make sure in the Chrome extension that the event sent is sent correctly and matches Coveo, schemas. So here we're really shortening the feedback loop that, that, that, that she's happening during implementation again for just saving time and going quicker into production. And last but not least, let's now focus on one major improvement we made to facilitate multiple use case and team collaboration at Coveo. So what happens is that many of our customer don't use Coveo just for one single use case. So you might be using Coveo for website, but you might might be also, users for commerce use cases, workplace use cases, and which mean that potentially there are multiple teams collaborating on the same platform and that they are resources that are relevant only for certain teams and not for others. And we definitely wanted to streamline that as well. And that is why we built this project feature. So what you can do with this product feature is it allows you to group your resources per use case. So let's say you have, again, you might have three different websites and and, and and also probably a a storefront experience for a commerce experience. Then you might be able to create four different project in Coveo. And on each project, you will associate only the relevant resources. Like for example, the sources, the query pipelines, the ML model, the reports, and tie it all under one roof, being a project. Once you have done that, you will be able to filter your whole, Coveo experience. So it means that you will be able, if you want to, you will be able to see only the resources for the specific product that you are dealing with. So it means less noise. You see only what is relevant for you as a user and also less, risk of, making a modification on the resource that doesn't belong to your use case. So here again, we just made customer experience much easier for this large multi use cases implementation. The feature is is being progressively roll, rolled out. So in you might not have access yet. In case you want early access, please reach out to your, CSM, and we'll be glad to activate it for your org. With that, I'll hand it over to Kirian. Thank you. Thank you so much, Michelle. I'm really excited about all of these new innovations. Hopefully, you're all also excited to start testing everything out. We'll stay a little bit to see if there are any questions. And if we don't have time to get, back to anything, don't worry. We will definitely follow-up with you after. So I see that, Am I, you're you're you're answering the first question. I might take the the second one. Will we be able to use a project to feature the usage data for targeted reporting? So you you will be able add so maybe one thing to keep in mind is this project feature will have a lot of follow-up improvements. Let's say that we're starting small and improving it as we go, and there will be definitely more follow-up on this. At the moment, what it allows you to do is to include only certain reports, in in a specific project. So instead of going through, for example, hundreds of reports in your organization, you might be able to see only the three reports that are relevant for your use case. So if if if just in terms of usage data, if you're, wondering, like, if if we're able to isolate the the usage from a specific product, at the moment, this won't be possible. But this is definitely one of the options we have for, for the next improvements. And that's why as well, we're, like, definitely looking to have more clients using the feature because we are basing ourselves on the feedback of these early testers to drive our roadmap and our list of improvements. So, of course, if if you think that this feature can help you, we'll we'll be glad to give you access and and and have a feedback loop with you to to drive the the next improvements. Great. Thank you so much. There's also a question around follow-up, questions for RJ and when this will be released. This is early access, so we'll be sure to follow-up with you to provide sir, further details of how we can get you, testing that out. And I believe that is all for now. As you know, this is part of, experience, sorry, Relevance three sixty. So this is part of four dedicated product sessions that are tailored to your use case. So if ecommerce or workplace are also related to your work, your use case, be sure to check those out. For any questions that we are not able to follow-up with, we will definitely reach out to you. Thank you so much for joining, and have a great day.