Hello. Welcome. We will get started in just a couple of minutes, so we can let others join. Stan, how are you doing today? Good. Good. Just great. It's a nice day here in my part of the world. So looking forward to getting outside, turning in. You're in Boston. Right? Yes. It's nice here in Connecticut too. Okay. Right. We're just gonna give just another minute or so to allow others to join in. Great. Okay. Why don't we get started? Thank you. Good afternoon. Good morning based off of where you're joining us from. My name is Juanita Ogin. I lead product marketing for one of our divisions at Coveo. Today, I am joined by my colleague, Stan Schroeder, who is a solution engineer at Coveo. Stan, are you ready to jump into our presentation for today? Absolutely. Let's go go for it. Awesome. Thank you. So for those of you just joining us, we will be taking you through uncovering total experience AI. How a total experience AI platform, which includes generative answering, can help your teams achieve more with less. So let's jump right in, and let's get started with the stat. It's always good to put things into perspective, and so When we think about why total experience is important, what you see here is is a statistic from Gartner that says, organizations providing a total experience will outperform competitors by twenty five percent in satisfaction metrics for both CX and EX. Now twenty five percent is a pretty significant number if you apply that to any base. Sign, whether that's revenue cost or satisfaction. But what I think is also important when it comes to total experience, it's being able to think holistically about your business, trying to reduce duplication or redundancies and trying to achieve more economies of scale for your team so that not only do they do more with less, but they get more with less, whether that's less technology, less processes, and sometimes more, you know, more smaller teams. I wanted to kind of cover these different definitions in the market around total experience. So again, here with Gartner, we can see that they define total experience as a strategy that creates superior shared experiences by interlinking multi experience, user experience, customer experience, and employee experience. So a nice holistic view, we would certainly agree with this. If we take a look at another analyst forrester, they have somewhat of a different terminology. They refer to it as experience architecture. But Forrester defines it as an approach that aligns the enterprise with omnichannel customer experiences, enabled by customer centric business practices, adaptable platforms, and connected employees and partners. And they see and experience architecture as being able to serve customers in the moment they choose, enabling employee to deliver great experiences, delivering seamless experiences across the customer life cycle. And sustaining beyond the tenure of executives, vendors, and technologies. And to be quite honest, I actually like the last point that they have here. Think it's important to almost have that future proofing mindset as well. Next, I'll take you into how we think about total experience. And there's a lot on this slide, so I'll break it down for you. But to think about total experience AI from the Coveo perspective, we do approach it from a technology angle because we are a technology company and offer a technology platform. So, we see total experience AI as one AI platform that can be utilized to deliver world class digital experiences across internal and external use cases and it gives business and IT teams insight into a user's digital journey while deploying cutting edge AI including generative answering to meet the constantly changing experience expectations that we know and we see every day. And what you see here is just an illustrated view of an enterprise which is made up of a collection of departments, each with their own systems and application, and within those systems, the underlying different data and format types, of course, underlying all of this are the IT and technology teams acting as a shared service to make teams productive and effective in their day to day work environment. And so if we go a level deeper into total experience AI in the way that we think about it, we can dive deeper into each of the various areas. For example, under employee experience, we've helped many customers improve different use cases within their organization. Everything from improving the intranet and search efficiency to making HR IT portals more effective even helping to enhance sales CRM's marketing digital asset management tools and, of course, helping teams like R and D and legal find and discover information more efficiently. On the customer experience side, we know when we understand there's a lot of sites that are out there that can and should be enhanced with things like unified search and AI, everything from your dot com sites to your support portals, community, partner, and training sites, and perhaps even service management applications. And when it comes to the IT side, here we're really focused on equipping and ensuring IT teams have the best of breed technologies available to them. Everything from APIs, to cloud native, headless frameworks and composability to microservices, integrations out of the box connectors, of course, enterprise grade security and AI, which I'll cover a little bit more in detail. But we also believe and understand that It's important to remain tech agnostic. Every enterprise is very complex, has its own systems and platforms that have been invested in over time, and so our goal is to be able to provide enhanced experiences across NEY any channel, whether that's your dot com, your chat bots, your mobile, and many of our customers actually use CoveIL within their existing platforms, whether that's the CRM, service management application, digital experience platforms, and so on. So, you know, we have a lot to offer and a pretty holistic perspective when it comes to total experience And what you can see here is our dedicated focus on innovation. We started back in two thousand and five with keyword search and have continued from there exploring things like Federated search to unified search, we began our investment in AI back in twenty ten, and since then have built many different models to solve for the various needs for our different customers, whether that's on the EX side, the CX side, and even the IT side as well. So what I'm gonna take you through pretty quickly because there's a lot here is a view of our different AI machine learning models that exist today. So these are real, they exist, they're in place, many of our customers are using them across different areas of their business. So let's get started. I have a couple slides here for you. The first thing I want to start with though is unified search, which is not an AI model, but which is the basis for providing relevancy to these models. And so, with Coveo, we help to provide unified access to enterprise wide information no matter where your users are searching, and we do that while respecting the permissions of your existing data sources. And even before AI models are applied, we have seventeen ranking factors and relevancy factors that we're applying. So again, that it's preparing these models with good information that they're going to continue to learn from. Right. So now let's get into these different AI models. And they're really ordered in terms of basic to advance, and almost represents our own customer's adoption journey along the way. So the first is query suggest. We've all typed into a search engine and if you recall getting suggestions to complete your query, that would be a machine learning model query suggests working for you. Once you type in your query, you're gonna get results. So the automatic relevance tuning model is ensuring that the best most relevant result is at the top of the list for you. And it's based off of the success of other similar users and searches, so that model's underway. Next, again, we've all searched before and often you get facets or filters on the left hand side aim to help you find information more quickly. Our dynamic navigation experience model is there and helping to reorder facets and sometimes auto selecting details for you, again, because it knows and it's it understands what other similar users have found to be successful. So those first three models are actually based more on you know, searching, actually a user, end user searching, where number four and five here don't require a user to search all. So we have content recommendations and product recommendations. And with these, effectively these machine learning models are making recommendations, whether that's content or products based off of who you are as a user, your profile details. And again, it's not requiring you to search, it is proactive recommending something that will be useful to you to resolve whether it's a product you're trying to buy or perhaps an issue you're looking to resolve. That's these models working for you. Now getting into our second set up here. Let's start with smart snippets. So, again, if you've ever searched and you've got a response, let's say a paragraph answer to your question, that is a smart snippets model or question answering working for you. And this is really great because it does not require a user to continue to click to find more information, the direct answer is being displayed automatically. Next, we have case classification. So for your existing customers who are authenticated or not authenticated, let's say they're looking for support they want to submit a case or a ticket. And what's happening here is our case classification models are helping the user self classify what that case or topic category or topic is. One, so they are able to feel more confident that you understand the issue at hand, but also to ensure that outing is done properly on the back end once the agent or customer service rep is receiving that ticket. And the next three models really are more on the commerce side, commerce oriented. They leverage our product embeddings and vector And essentially, we're using more intent and providing more personalized recommendations on product ranking, query suggests, so as you're looking for products or product recommendations. This is all based off of the users in session shopping behaviors and actions, these models are working again to get that user to the right product, which will eventually lead to more revenues to the company. Last but not least, we have our Coveo relevance generative answering product, which is leveraging the latest in AI, where we're taking multiple different documents, passages, we are applying our relevancy and tuning. We are then sending them to LLMs and generative AI, to come back to create a response for the user's need at hand. So this is the generative AI, component and response that many of us are hearing about in the market. So hopefully this gives you an idea of the breath and the different use cases that these models were created to solve for, and it's important to understand that because we're all familiar with a buyer journey and the fact that the buyer journey isn't linear, right? You may have an unknown customer that's clicking through your different sites, at some point they may decide to log in, and so you want to really be able to capture, capture users and deliver them relevant information no matter where they are. And so what you can see here is just our view of the buyer journey, the presales, to during sales and post sales, the different conversion metrics and KPIs that you can measure along the way, but also it's important to understand that some of your sites for your customers as well as your employees correspond to these different stages of the buyer journey as well. And again, the idea here is to ensure that your end users are getting what they need no matter where they are. Now, before we get into the demo and before I pass it over to Stan, I wanted to end on connecting the dots and user tracking, right? So your users are jumping from site to site, application to application, How can you track and gain insights from their activity? Well, if we use the same buyer journey model and look at it from a state's perspective, so unknown user to known and authenticated, then we can draw a lot of insights with our tooling. The first is Coveo understands the importance privacy, so we are deprecating the use of third party cookies and only leveraging first party cookie tracking only. On the unknown side, we can help because we do we do perform anonymous user tracking within the same top level domains. As your user goes through their journey, we actually build a user profile for them. And that gets used in places like our CoveIL for Salesforce action component, which you'll see shortly. And the user profiles also used by machine learning to know what to recommend and personalized to end users across their journey. Some of the user dimension information that we'd be leveraging there are history of a user's actions, topics, their favorite brands, favorite categories. Now your users continuing along their journey, perhaps they're authenticating, and so we are performing user stitching, including how they switch from one browser to another, how they move from unknown to known and authenticated, and even if they're accessing multiple accounts across different interfaces within the same browser. Lastly, we do provide you the ability to send along custom contextual information, allowing us to build fuller, more broader understanding of the user and what would be relevant to them. So I hope that gives you an idea of we think about total experience and how we're enabling companies to achieve and deliver upon this holistically. So now we're gonna get into the demo portion, and Stan is gonna take you through six different use cases or six different new eyes that demonstrate the total experience and practice, and you'll get to see many of the AI models that I mentioned. Sastan, I'm gonna pass it over to him. Great. Thank you very much, Juanita. That was well done. Okay. And you can see my screen? You're all good. Yes. Alrighty. Great. So one of the core principles of Coveo that makes it so valuable for this idea of a total experience across self-service service in a variety of other applications is that the Coveo index is designed from the ground up to be used for multiple use cases, audiences, and multiple search pages at the same time. Some of the reasons why we can do this so well is because of The following. First, Coveo index supports a secure permission based search, or in other words, early binding security. Meaning, that Only someone that is authorized to see a given document or knowledge based article, etcetera, will see that article. Second, the Covayo Index is also uses this concept called query pipelines, where you can design different relevance rules for different search pages or audiences or use cases. Thirdly, the Coveo index is built on a world class cloud infrastructure that automatically handles scalability, reliability, and availability all at the enterprise scale. We've got many examples of customers that do this. One example is actually Salesforce. Salesforce uses Coveo in many different parts of the business both internally and externally facing. This is the Salesforce help and training website. And everything that you see on this site is Salesforce except for the search bar. That's Coveo. And, obviously, we all know that Salesforce has thousands of customers across hundreds of countries all over the world. They service these customers in over a dozen languages with Coveo. Coveo has been deployed in more than twenty areas of Salesforce's business, including what you see here, the help and training website Let me go ahead and do a search here. In addition to the help and training website, Another example is the Salesforce AppExchange that you see here. This is also being powered by Coveo. Another key aspect of Kovea was that the system does recognize the ability when a user traverses multiple different search interfaces. As one you described, we call this user stitch So, Coveo tracks who the user is, even if they are anonymous. So, for example, if the user comes to your website, does a search, as an anonymous person. And then later on they authenticate maybe to create a, you know, a case or to make a purchase or something like that. Boveo can track that user and associate both user journeys to the same person. So let's take a look and see how this looks in practice. So let me introduce you to Barqa. Barqa is our demonstration company. So, they are in the Pleasurecraft manufacturing industry. So, with customers and employees located in multiple countries all over the world. We're gonna start here on the BARka support site because our customer, John, is having a problem with the Barca Skipper charts and GPS app that he downloaded. Notice when we start on the homepage, John's experience is fairly generic. The Coveo recommendations will automatically get more personalized as John searches and clicks around. But for now let's go ahead and log in and help John solve this issue. So, let me go ahead and log in as John. And, You can see here the recommendations are now personalized to John and the products that he has, he owns specifically and is registered with us. Now, I'm gonna do a search here, and I'm gonna say, I'm having trouble downloading Math updates. There we go. This is my query suggests. Now, The nice thing about this query is that it's a great example to see the different capabilities at work here. The first you'll see is the unified search results. So, all the different content sources across the top here. He can search across product information, knowledge base, information, closed cases, downloads, unstructured community posts, etcetera. He doesn't have to go to different locations, one place to go and search for documentation, and another place to go and search for the knowledge base for support. It's all right here at his fingertips. The next element that we see here is the smart snippets, though as we'll need to describe smart snippets as the large language model response, where we're giving John a direct answer to his question. Saving him time, making it really, really easy and effortless on his part to find the answer to his question. The next element that's going on here that's helping people like John, as I scroll down, you see these recommended flags this is that automated relevance tuning, machine learning model at work. So, basically, what's happening here is other people like John, when they've had this question. These are the the solutions that most frequently, they have found most helpful or or lead to a a positive outcome. And so the the engine is looking at those behaviors and elevating these solutions towards the top of the list. Now, the other, the next element that you see here is the query suggests, so you noticed, as I was typing in my question, The system is responding by making recommended query suggestions. These are really important because what this does is this helps people form better queries. And so when, especially, you know, when people are not as adept with things like, you know, just grammar and spelling etcetera. BRI suggestions are a big advantage. And this is an ML driven model. It's not just statistical quarry suggestions or the fact that we get these frequently, these query suggestions come up because people like John meaning his, his profile characteristics have asked these questions and have had a positive outcome The next element here, as what I need to describe is dynamic navigation Dynamic navigation is an ML model that applies the learning to the filters on the left hand side. And so it automatically ranks and adjust the filters that are occur because of this query. For users like John, again, based on patterns of behavior. And they can they can even get intelligent enough to preselect filters and facets as well. So with this one query, there was no less than four different ML models at work. Assisting John and finding exactly the right content to solve his problem. All at the same time, all pre built. We've done all the data science for you. Takes minutes to deploy, and literally it's just, you know, a few clicks for each model to deploy them. Now, for the sake of the demonstration, let's say that John didn't go and do self-service, but instead decided to go right to the support page for help. So let's help follow John as he creates a case with Barca to help with his GPS app issue. So, I'm going to click on contact support, and we're going to get the opportunity here to create a case. And now, I'm going to go ahead and start describing my problem as John. We'll say the GPS trip option. Is not working. And you'll notice that as I'm typing in my query, The interface is changing. What's happening here is another machine learning model is at work, automatically characterizing or categorizing this case as it's being created. So the reason why we do this is that we want to make sure that the the case is tagged correctly so that it gets routed to the right folks to respond to his question. So rather than it getting routed to, like, the parts counter or, you know, the group that helps people install or set up, you know, radios or etcetera, we want this to go to the right folks that can help John solve his problem. This is helpful to barca because it saves time and money around rerouting of cases This helps John because he's going to end up getting in touch with an expert faster that's gonna help him solve his problem. Quicker. Now, when John hits the next button, the next step here is why don't we see if we can help him resolve his problem without escalating this to the the tech support center. And you can see here we're gonna make some different suggestions here on different possible solutions to his problem. Now, at this point, John can click on these. He can If he resolves this problem and cancels out of this, we'll record that information. If he clicks on some of these documents and goes ahead and creates that case, we're gonna record that information as well. In both of those behaviors, positive and negative, we'll influence the the learning models going forward. Now, at this point, John would submit his case. It would now get routed, and This is actually a place where you'll see that user stitching capability that we talked about in action. And that's actually a good time to transition to the role of the customer service representative that's gonna assist John in resolving this case. Let me go ahead, and I'm gonna change roles here. For just a minute. Now, let's say I'm Marie, and I'm a customer service representative of Barca, and She works with specifically with our GPS and charting app customers. And let's see how Coveo assists her in her job and she assists John with the case that he's created. Now, the same index that we deployed on the Barker website search, in the barcode support portal is also used by the barcode customer service representatives. But this time, it's fully integrated into their Salesforce cloud. Now, as I open up John's case, you'll notice that the Coveo insight panel has launched here in the context of this case on the right hand side. But in it's not what we've done here is we've replaced the standard Salesforce knowledge panel. But Salesforce knowledge is still an important part of the insight panel. But instead of just being limited to Salesforce knowledge, John also has access to all of the different sources inside of Barca that he might go to to help people resolve issues. Though this could be documentation. This could be, you know, closed cases. It could be community posts. Downloads. It could be other resources such as, you know, Jive, Jira, confluence, or SharePoint, all now available to Marie as she's helping the customer. The whole time, though, she's she's able now to stay inside of Salesforce. Without having to, you know, alt tab to other systems to do different searches. Unified Index, unified search response. The reason why we can do all this is because Coveo supports that idea of early binding security or permission based searching. And so we can be sure that that the customer won't see information that they're not supposed to see and Marie will only see information that she has been permission to be able to see. You'll notice too that when we launched the COVID-site panel, it's automatically started searching. We've used the context of the case to launch the panel. Now the context of the case is configurable. The standard context is typically the subject and the description fields. We can pass any other fields that make sense as part of that context, so that Covea was searching intelligently right from the get go. So, it could be product you know, version information. It could be platform information, etcetera, whatever makes sense. These fields can easily be configured inside of Salesforce, and these fields can be changed as your business grows or changes. And when all this information is passed over to Coveo. Coveo was searching across all those different sources to find the best answer. You'll notice at the top of the list here, the smart smart snippets are in effect, so the same capability that we saw out on the website The same machine learning model is is helping John is helping Marie solve John's problems. Similarly, the automated relevance tuning, these recommended flags, those are the the same model is in effect here as well. And, where we suggest are also working here, as well as dynamic navigation So the same capabilities that we saw, John taking advantage of out on the website, are helping Marie as well. And the benefit to doing it this way is that the system is combining the learning from both audiences. So the same learning that it's it's getting from Marie's actions is influencing the behaviors on on John's side of of the view and vice versa. So we're taking advantage of learning across all of those different interactions. Now, we mentioned the user stitching before. Now, look what happens when I click on the customer actions button. So this is a view for Marie that shows her what John was doing just before and as he created the case. So what this means is that she has some insight as to what products or what documents he's looked at, what kinds of queries he's performed, recently, so that she can be more personal and relevant to John. So that's that idea that user stitching in effect here really clearly. In addition to what you see here, Marie has some other actions she can take. So, for example, she can preview a document without leaving Salesforce She can send a link to John with a solution here. She can send this as an email message. She could post it to his case feed. And then finally, if she, you know, decides that this is a knowledge based article, is going to be the one that she's gonna use to solve the case. She can attach that to the case so that now we can see all the different ways, all the different solutions that were used to resolve this case. This is useful so that if John calls back with another problem or with the same problem, the next person that gets this can see how Marie resolved -- help him resolve his issues. Likewise, we can use this for reporting purposes so we know what solutions were used to resolve what cases. And all these actions are now tracked by Covalent and influence those models going forward. But what about your other employees? So what about those employees that are not inside of Salesforce or not working in the call center using the sales or service cloud? How can we help those employees as well? So, Coveo is going to answer there as well. We call that Coveo for workplaces. Again, the reason why we can support this idea is because that secure searching, permission based searching capability that I referred to earlier When we index the content from the variety of different sources, we're also indexing the security model that's native to each one of those sources. And building that right into Coveo. So, it's fast and reliable. Coveo for workplace can be integrated or deployed into lots of different applications or environments. It can be integrated into an existing Internet site. Or it can be integrated into an employee experience platform or portal. Or like you see here, We're demonstrating a standalone purpose built search page around the idea of employee enablement. We call this the Barker Group Workhub, and this is where any of the employees from Bark can go to search across all the different enterprise resources that they need to do their job. One of the, the elements, important elements of this particular interface is that we've adopted the idea of consumerization of IP. So the design elements of this search page embrace this concept Consumerization of IT has been described as the embrace of consumer like online services in the workplace. And as an example, I'm gonna go to one of our customers. So this is Bass Pro Shop here, and this is Coveo powered search interface. And you can see as I clicked in and started searching for ply fishing, I'm getting immediate search results here on the right hand side and suggested keywords here or query suggestions here as well. So it's it's we take the same idea and we apply it to our Covea Workhub. So when I click into the search bar, You can see here, as I start to type in my query, I'm getting query suggestions here in immediate search results here on the right hand side. So the same capabilities around automated relevance tuning and query suggest are available here. This is just a nicer, more effective interface. Another aspect of consumerization of IP is that idea of recommendations. So, you see these popular with your colleagues and recommended videos. You see that kind of content and recommendations out on customer facing websites. Why not do the same thing for your your employees as well? And these recommendations are based on yet another machine learning model provided by Coveo. So, the recommendation engine is automatically personalized and automatically updates and gets more relevant based on users' profile characteristics, as well as their patterns of behavior as well. Another element of personalization here. Let me go ahead, and I'm gonna do a search here. And now I'm searching as Steve, who is a U. S.-based employee. I'm searching for the term pay slip. But and you can see here, I am getting US focused search results. Now, I'm going to change roles for just a minute. I'm going to change from Steve to one of our other employees, Jesse. Jesse is based in Canada. So rather than being a US based employee doing a search for payslip, I'm searching for payslip as a Canadian employee, and you can see here the results that are coming up are to my Canadian colleagues. So, that's another element of consumerization ID. The personalization asked the search results that we've embraced in this, in this search interface. Another capability here that's really helpful and specific to workplace environments, if we get this, you know, request a lot from from our customers, is the ability to do people searching inside of Coveo. And you can see here, we're actually started that right here on the search interface as recommendations. So there are two recommendations in effect here on the homepage. Peer influencers. So because Jesse is a manager, the peer influencers is a recommendation based on other managers. So these people represent the go to managers that other people like Jesse access or look to for help and guidance. Topic influencers is based on Jesse's specialty. So because he's a security and safety, focused manager. These are other managers or other people that have that in their specialty title. So two slightly different recommendations, but really important recommendations here. For people search. Now, let me go ahead in, and let's say I'm looking to speak with training manager from our WaterCraft division. And I can see my search results. Click over to, well, there you get results right away. But let's click over into the crew members, and you can see how easily I can find the the type of people I'm looking for with the appropriate filters and facets here on the right hand side. Now all of this leads to another sort of a thing that's getting a lot of attention and excitement in the press lately, talking about generative AI and Chat EBT. So, generative, as we all know, generative AI engines are already being embraced in a variety of scenarios out on the web. As a matter of fact, it was interesting. The one of the first times I heard about generative AI from a non search professional was actually one of my kids who was using it in college. But the whole topic of generative AI opens up a lot of problems with the enterprise. And and many of you might have seen the news stories about organizations losing control of their own intellectual property when using generative AI and Chat BT. So, this opens up a question. How does industry embrace the use of Chat GPT and generative large language models as a way to increase employee efficiency, but at the same time guard against the loss intellectual property and security. Well, Coveo is going to answer for that. The Coveo approach merges the two ideas of search and generative experience together so that rather than having to go to two different places to ask a question, and possibly getting two different responses, we've merged both together at the same time. The next part of the Coveo approach is security is an absolute must. You do not have to send your complete document library to a third party large language model that you don't control or own. Another element of the Coveo approach is that we recognize that not every interaction requires the high cost of a generated answer. Sometimes people will really just search and not looking for a synthesized answer. They're looking for a PowerPoint presentation or they're looking for a specific document. And hosting your own LLM is often a very, very resource intensive and expensive proposition. So Coveo combines the best of both worlds. We're going to have webinars that demonstrate these capabilities with generative in more depth. But I did want to give you a quick example of what it looks like. So I'm going to ask a complex question here that's got multiple parts. And I know that the answer is located in multiple places on our website. So, I'm gonna ask a question. What's the difference between a hosted search interface and a standalone search interface and how do I create one? And so now what's happening is Coveo is providing both the search results that you see along the bottom, but also generated answer at the top of the list. And I'm going to go ahead and I could reformulate the answer as a bulleted list. Likewise, I could reformulate the answer as a series of step by step instructions. This is just quick example of the approach that we're taking. The final release is also going to include citations, so It'll tell you exactly which documents or which web pages it used to generate the answer. And then it'll also enable you to interact with the model as well, asking follow-up questions, and that sort of thing. So, just a quick example of what that looks like. Now, the next element of Coveo that's -- I wanted to end things with is from a demonstration perspective is the idea of analytics. So, how do we know if what we've deployed is actually improving things for our our end user community or our customers and our employees and our partners. So Coveo has built into the system a porting and usage analytics capability. And this, this report, this health health check tab is a really, really good way to get an idea of how successful we are at helping people resolve their issue use. What's nice about this, this tab is that beyond just the information, it's actually got some descriptions here that describe actually, you know, what the numbers mean and how to interpret them. But these include the visit click through rates, you know, out of all the different people that came to your site, what percentage of them actually clicked through and looked at a document versus the search or query click through rate, which is how many people that actually searched or performed some kind of query went ahead and clicked through a document. The average click rank in the content gap. So click rank is, you know, which document did they click on most of the time? Is it at the top of the list, at the bottom of the list, on the second page, that the, you know, the third item on the second page, etcetera. So we're recording and looking at the trends associated with click rank. Which is a really important measure of success, as well as the content gap. So out of all of these, numbers, the first three numbers, we want to see, you know, trends that are sort of going up. We want to increase click to rate, increase click rank. And likewise, with the content gap, we want to see that reduced. So, we want to see fewer and fewer content gaps in your in your user's search behavior. Coveo has a whole number of different pre built and use case specific analytics, such as self-service, case deflection, case assist, commerce, and many others. We won't have time to go into all of them here, but I'm going to turn things now back over to to Juanita, so she can talk about the business outcomes associated with the total AI experience. So take it away, Juanita. Think Sand. That was a great demo, a lot of different use cases. So, as Stan mentioned, he showed you some of our in product analytics, and sometimes we're able to actually show you cost savings and revenue influence that are different products and models are delivering. But other times, we know that business outcomes and ROI happens outside of the product. And so we work hand in hand with you to understand what the value is that we are bringing and realizing through the use of our platform. And we're able to do that and demonstrate that across different use cases. So what you see here is the different range of improvements we've helped our different customers achieve within their service and support use cases, website, workplace, and ecommerce use cases, a wide variety of different improvement metrics that are real and our customers are benefiting from on a daily basis. We hope that you enjoyed what you saw today, and we'd love to help get you started down your total experience journey. If you like what you saw and you'd like to learn a little bit more, we invite you to book a one on one discovery call with us. We will break down this total experience category and focus on a specific use case that you can get started with. If you're not ready but you wanna learn a little bit more, I do invite you to watch Arjan AI demo where you can see how we think about generative AI, our response, generative answering to it, and how we are approaching the next evolution evolution of search in AI. With that, Stan and I would love to thank you for your time and attention. Please feel free to connect with us and thank you. I know we were a bit at time, but I did wanna just offer an opportunity for any q and a. So I'll give you all a moment just to send any our way. Stan, I do have some that did come in. And I think it's best for me to push these your way. Okay. Sure. The first one is the workplace experience employee hub that you showed, is that an out of the box template? So actually no, it's not. It's sort of an art of the possible. It's based or built around our headless React library. It's not hard to assemble, but the idea is to convey a bunch of conceptual ideas. We can provide it as a sort of a starting point. But it isn't a, you know, a prioritized template. It's absolutely available for reference and for learning. And it's certainly popular as we've seen with other companies. The next question I'll ask is, The question is, do I get all of the different AI models covered by purchasing the Coveo platform? All pitch that one to you. So Yeah, yeah, absolutely. So all the models are available out of the box. Other than generative, which is we haven't released that yet. So, don't know exactly the details yet around licensing per generative, but all of the other models They are available out of the box. And once you have licensed Coveo, you can deploy those mix and match the way that you need them. I absolutely encourage trying them out and exploring what they can do for you. And if I could add some models are extensions or use case specific, so you might buy Coveo as a platform and you'll get many out of the box some are more specialized like the commerce ones or the -- Right. -- these classification ones that would be like an added extension or package to purchase, but Yes. Many of them come with the basic Quail platform. Last question here is How long does it take for the machine learning to start working? Yeah, so great question. Well, before machine learning in, you're going to have great search. Because you can, with Koveo, when you create your query pipeline, you're setting up the basic Relevance rules, Coveo does great searching out of the box. ML will kick in as it acquires enough activity Typically, we sort of generalize and say around ten thousand events, events being searches, you know, clicks, refining searches, etcetera, those, all those events in total. At around ten thousand events, it'll start kicking in. And what we found typically is that through user acceptance testing as part of the deployment. You you kind of you end up sort of getting almost enough of those events for the AI engines to start kicking in quickly. But roughly, it's around ten thousand events. If you've got a really active website, for example, you're going to see it start to work very, very quickly. Awesome. Thank you. While I think we're out of time, we've gone a little over so Stan. Thank you so much for that folks. That extensive demo. I think it really illustrates total experience and practice. So thank you all again, and thank you, Stan. Find now.