And it's cicholic. Cicolic. I told you I'd remind you. So we're just gonna give everyone a few seconds to jump in, then we'll get started. Okay. So we'll get this started. Thank you everyone for joining this webinar. My name is Lydia Ziau, and I'll be your moderator today. I'm really excited to be part of this of today's session with Colin Strachan, product marketing manager at Coveo and Isaac Sacolick, president and founder of STAR CIO. They'll be covering how IT leaders are tackling the challenges of the digital and hybrid workplace in twenty twenty two. I have a couple of housekeeping items to cover quickly before we get started. So first, everyone is in listen only mode. However, we do want to hear from you during the today's presentation. We'll be answering questions at the end of the session, so please feel free to send those along using the q and a section on your screen. Today's webinar is being recorded, and you'll receive a presentation within about twenty four hours of the conclusion of the event. On that note, I'll leave the floor to Colin and Isaac. Thank you very much, Olivia. Thank you everyone for joining. So my name is Colin. I do product marketing for Coveo for Workplace. And today, we're gonna be talking about what you can do with AI powered search to improve employee experiences. And it's a real privilege to be joined today by Isaac Sikolic, who has had a very comprehensive career in digital transformation. He's held multiple CIO positions. He's a published author and industry speaker, and he's also the founder of StarCIO, a digital transformation consulting company. So thank you for joining us, Isaac. How are you doing? Thank you. I'm doing great, and thank you for having me, and greetings, everyone. Fantastic. So Isaac's gonna share some of his experiences in his research today, but just to kind of frame this discussion. And We recently became a public company, so I I need to show this disclaimer. If you would like to inspect it, you can obviously, pause the recording after the webinar and take a look, but it's really just a legal disclaimer around the contact of this presentation. So I think we all know that the increased digitization of the workplace has introduced some challenges to employees. And every year, Coveo conducts market research to get an understanding of how people are feeling about the standard of digital experiences across our solution areas, which means ecommerce, customer service, and, of course, the workplace, which is what we're talking about today. So we're gonna be publishing this report over the next few weeks. But to give you a sneak peek, so one of the kind of bigger findings of our report is that sixty percent of employees are searching within four or more data sources every day, and eighteen percent of people are actually searching within seven or more. So that gives an example of this kind of digital sprawl that's developed in the workplace. And because of these data silos, you know, employees often don't even know where to start looking. And you could see the impact that this can have not only on the level of confidence in their work, but even on their on their well-being. So what I'm gonna do is I'm gonna throw this over to you now, Isaac. I mean, why why are we talking about employee experience today, and and why has this topic become ever more important as the workplace becomes more digitized and more distributed? Yeah. I I think, you know, look, we're coming off of COVID and remote working. We've experienced, the loss of the water cooler, for people to find people and ask information from them, in hybrid environments. And now with the great resignation, we know people, are considering places where they feel they're valued, but places where they can be effective and productive. And, a big part of corporations responding to that is looking at the employee experience, much like they've looked at customer experiences over the last five years in their digital transformation programs. Now, when you use the word employee experience, it means a lot of different things to different people. For people in IT, they still think predominantly that employee experience is around the devices and user computing, mobile devices. Do I just have access to the network and things like that? And if you ask HR, you know, HR is gonna talk about their portals and access to HR information and benefits, and things like that. But when you ask an employee, it really comes down to some of the stats you just shared, right? They're trying to do their work. They're trying to figure out how to procure something or who's an expert in their organization or what's the business process that they have to interface with when working on something. Maybe they're a new developer who's come and joined the organization has no idea what the code they're working with is in front of them. And so there's a huge productivity factor in terms of being able to find information effectively. And there's also a huge, important factor in terms of helping people feel like they're not wasting their time, but they're not spending a lot of time looking for information that should really be at their fingertips. So when I look at the definition, what I think of employee experience, I think about how organizations make it easy or in many cases, not so easy for people to get work done productively and easily. And a lot of that comes down to looking at the entire ecosystem and information that's available to employees and how do you make it seamless for them to find the relevant information for them to do their job? Okay. And, you know, what you said about productivity there, it's interesting. It reminded me of, you know, the American educator Peter Drucker. One of the famous things he said that became quite a famous quote before he died was that, one of the most significant contributions that management can make in the twenty first century is to increase the productivity of knowledge workers. And, I mean, I think we know that the increasing productivity is about giving them access to information. So, I mean, if access to information is so fundamentally important to to to powering that and to increasing productivity, then why are we here, you know, twenty years later after the first search indexes were introduced, talking about this? You know, why are we still not succeeding in getting that information to employees? Well, I I think it's because we've looked at the problem from the perspective what some of the earlier technologies could do for us. Right? When you talked about search, you know, ten, fifteen years ago, we were doing things, like keyword searches, indexes, essentially built off of document stores, so where we were storing our word documents and things like that. We've been doing it for a decade or two. So, you know, the fact that we have really new documents in there, we have versions of documents in there that are hard to parse through. We have really old documents in there, and that's that's essentially what employees see when they do a vanilla search as they start typing keywords in and they start getting a list of documents out. They're not really sure what to do with that. We think about the other use cases around search. A big part of it is around customer facing capabilities. And, you know, those are search engines that Barca, again, pre, constructed with information that, is of, you know, that we wanna make available to customers, but we're not necessarily including profile information and segment information, shopping information, things that they're using in the course of their activity on websites that we have that are really important context, for search to be able to deliver a good experience. When I think about customer service, right? I talked to CIOs, for example, about, you know, what tools do you have in place for customer service? They named two, right? They named the ticketing system for customer service and then access to the CRM. Now, somewhere in here is all the vast information around the product itself, around how people are using it around. What's the last thing that a customer looked at before they called up customer support and all that's really important information. So we're dealing with the notion of what search was a decade ago, very siloed pinpointed point solutions against those three different problem sets. And now we think about and have expectation that a search can handle looking at all those different data sources, being able to do relevancy ranking, being able to do more than just, keyword searching, we're getting into natural language query searching, and then also starting to couple it with external data sources. Right? Well, how is the news? How's the weather? How's stock prices? How's what's happening in COVID and other political events? How's that impacting what sh what people will need to see early on? And you put that from the employee perspective. That's the kind of information when they say I'm looking at four or five different information sources, it's because they can't find it all that in one single source for themselves. Yeah. I I love that idea that, you know, of how intrinsically linked the employee experience, is to the to the customer experience. And I see a lot of people now talking about as opposed to CX and EX, they're talking about TX, the total experience. I think the holistic kinda connected journey is incredibly important. I mean, you've you've obviously seen a lot when it comes to digital transformation. I mean, you were CTO of a newspaper SaaS company in the early two thousands. You've held multiple CIO roles, for, I think, media, data, and financial services companies. What have you learned from all of those experiences, and and what stories can you share about about implementing the digital workplace specifically? Well, I mean, you know, search was foundational to how media companies delivered product back to their end consumers. And I helped newspapers going well back into the late nineties, to help them take, you know, their newspaper content and put that up on the web. And we had to spend a tremendous amount of work extracting searchable terms, you know, whether you call tags for categories or key value pairs, but we spent a lot of time pulling that information out. And then later on, we added information around semantics. When I worked in b two b situations, we had huge taxonomies of information that we had to marry, into our content stores. So think of an entire building product set of catalog and applying it to plans and specifications that go into a building plan or building specification and being able to search on that or rich information around financials around, different equities that we might be doing and trying to aggregate information on. And the problem was so difficult to marry the, you know, the taxonomy with the structured data, with the unstructured data that we ended up having to solve the problem in different silos. We had, you know, one group focused largely on customer experience, and building out the right sort of tools and technologies to deliver an experience there. And we did relatively little for the employees. And so you think about the employees, they couldn't find the same information that our, our, our, our customers were able to find. We were using lots of different technologies that we had to bring together and hamstrung together. And I think the most important thing from back those days is we had a significant gap between what subject matter experts understood about the underlying data and the content and the programming that was involved to make that, relevancy tuning, to be able to make, the right decisions about what to show up in search results, to figure out, what parameters to search on and how to use those parameters and finding results in. And so, you know, a lot of if you skip forward today, a lot of those technologies problems are solved. We've learned how to aggregate information. We learned how to auto find information in unstructured information or pulling in data in terms of usage. And now we have machine learning to be able to, look at usage factors and decide for ourselves what relevancy is rather than having experts come in. And so we had to do a lot of that by hand. I remember doing this at business week where we had to go back and prove prove to editors that a machine learning algorithm could outperform picking results out compared to what our editors are doing. I tell that story a couple of times because our editors were paid to be experts, in terms of what people are looking for. And here comes a machine learning algorithm, looking at the context of the search, looking at who the user is, looking at past performance and saying, look, we think we have a pretty good idea using math and tell people what to look for. So I think, you know, if you look at how executives today think about search, they're thinking about it from how they implemented search five, ten, fifteen years ago. Lots of work to build the data sources up, lots of work to present that information to an index, a lot of debate in terms of how to tune relevancy ranking, working between subject matter experts and, the technologists, a lot of work to just have the infrastructure running if you have huge scale in terms of data or in terms of usage. And so you have to put a lot of work in to build and you have to put a lot of work into support. And so, you know, part of what excites me about where searches today is you can get a lot of the heavy lifting out of cloud. You can get a lot of heavy lifting because search technologies like Coveo are marrying the, work to do, extraction off of unstructured data with the work to be able to search and do relevancy turning and then the work to enable, developing experiences during using a number of different formats. Okay. And you've talked about I mean, you talked you were talking there about the the amount of work being, you know, a challenge for for tech leaders. I know you've been speaking recently with several, you know, CIOs, CDOs. I know there's about their search implementations. What what are their pain points if you have you come across? Because, I mean, we we know what's frustrating employees, but in trying to tackle those challenges, what challenges are IT leaders facing? Yeah. So the research you're referring to is I spoke to five, leaders of technology and of product management and of data across different industries. And we just had a conversation about their priorities, and their perception of how search and how unstructured data fits their overall strategies. And there's a couple universal things that came out. I mean, first is this notion that, you know, we're everybody is doing digital transformation. I think that's pretty well publicized. I think all of them are focused on customer experience. They're focused on enabling data and analytics and data governance in their organizations. Some of them are doing it from the citizen data side. Some of them are very, strong at the machine learning and data science side. They're also modernizing workflow across their employees, you know, and so if you talk to them about what you're doing with their employees, a lot of it is around modernizing legacy systems, putting automation in and things like that. And so the, you know, this notion of what do you do with unstructured data and how does that fit into the strategy is coming out after those, those, those priorities. Now, if you look at me, you know, I like to leapfrog where people are today, and that means looking at where capabilities are today and where they need to be over the future. And so when I look at the customer experience and data side, I think there's vast opportunity to find, areas of their intellectual property in unstructured data and in third party external data that can bring value and be more competitive than just looking at their structured data. It can feed machine learning algorithms, it could be hyper relevant to helping analysts inside their organizations make better decisions over how to present customized information. That's happening a lot in financial services. So it should be part of the data strategy. It should be part of the customer experience strategy in terms of how to use that to competitive advantage. And then when you look at, you know, the impacts of automation, right, we're trying to automate things so that we can enable employees to do more, better decisions, faster decisions, smarter decisions. And a lot of that is going back to what you presented earlier when they have to find information to make a decision, to understand how to change, transform or change something. Can they find that information, easily at the fingertips? Can they find subject matter experts? Can they see how a process is working? And, that is still very hard for them to do. So I think it needs to be more cornerstone to their strategies to figure out how to find easy on ramps to bring the unstructured data world into their data worlds. K. And you you mentioned a couple of times about this idea of bringing external data, indexing external data. So do you have any examples of of where that's been deployed, where that's been that strategy has been employed and and and how it's helped employees to to make decisions? Yeah. I mean, I just think about, you know, in financial services, news is an early indicator that inequity is going to move, that there's a a major disruption happening in the marketplace, before markets respond to it. I, you know, so I think news in financial services has always been a competitive advantage for organizations to look at. I think about retail, right? You know, you think about shifting customer demand, for specific product lines beyond what intuition tells you about seasonal activities or what, you know, the analyst firms are gonna tell you there's gonna be huge interest in certain products. Being able to listen to social media and say, you know, there's a lot of people talking about this particular anime or this particular set of products, or maybe, an emerging sport that you didn't think of. You know, there's articles coming out last month that pickleball is now one of the top sports being played by boomers. All those can impact retail in terms of, of, of how they're presenting content, to their end users. Even manufacturers looking at supply chain risks and looking at, their product information when they're looking at how to outfit a product for a particular, client that's calling up to find information about their product offerings, recognizing what the weather patterns are in their space, recognizing what kinds of things are happening in their space that's driving demand and being able to print the right products for those customers, from their, their, from their vast catalog of product offerings. All those, you know, you have your database, you have your skews, you have your information sources, but without looking at how customers are actually using your product, that's, you know, one dark data area, many search engines, many data sources and warehouses have, and then looking at external data and saying, how is it influencing whether to dial up or dial down the ratings of specific results? I think that's a second area that that comes to play there. Okay. Wow. So that's really going from, you know, search removing bottlenecks to search really enhancing what is possible, with the inflow experience. So I I wanna ask a bit about AI then. I know you I know I know you touched upon it. And, obviously, AI has been touted as a solution to to many problems in the workplace and in in society in general. But what excites you about AI powered search specifically, and and and what do you see it doing, to improve the experience? Yeah. So I think let's just, you know, leave the word AI out for a second. I mean, I think what we're really talking about is getting under the hood and realizing relevancy is still a very challenging problem for most organizations. You throw, a lot of unstructured data. You know, maybe you get a a really well formed query against it, and you have a lot of information about the end user to set context. Maybe you don't, But you don't wanna be in that situation where I was ten years ago and trying to figure out and use subject matter experts and using tuning algorithms to figure out how to present results back to the end users. You wanna use past performance to do that. You wanna use machine learning to look at all the dimensions of data that you have and use that to con continually refine what to present back to users. So I think, you know, it's solving a huge gap between how to make sure that your results are not, you are not just relevant for what your subject matter experts and your end users are saying, but how it changes over time and over seasons and even getting into personalization. So I think that's, you know, a big part of it. I think a second part is, you know, again, a decade ago, you would be using different technologies to extract information out of documents, and then you would have to then present that information back to the search. Right? Search back then were basically indices because Oracle databases and SQL databases didn't handle well this notion of writing a query or searching an unstructured, data source. So indices were what they were essentially doing, but to present the information back to those indices, you had to do a lot of the, work upfront. And so now what you're seeing is technologies like Coveo are bundling that together. And that means I can use a search engine, and their APIs is an appliance to my analytics. Like, let it feed the machine learning that I'm doing. Let it be part of a distributed query that's between structured data sources and unstructured data sources. And that becomes really interesting to me. And then when I think about, where we're going, in terms of interfaces, you know, we all know keywords are a really difficult way of searching. So now we put keywords in with information about the user information about the, the past experiences, all the activity that they've been doing in past behaviors. Now I have a much wider dataset that I can use to form their query. And now I'm going to say, instead of using queries, I'm going to use natural language querying. I'm going to do natural language querying in that's multilingual. So I can allow a German or Japanese speaking person to write a full set of questions. And my engine is going to understand it. Where this is going is hopefully voice activated search. I think, you know, we're starting to see that as a possibility through, tools like Alexa. One of the people I interviewed mentioned hand gestures as a way of doing searches. And, you know, we talked we saw movies about that two decades ago, and I think that's gonna come very soon in terms of being able to look at vast information saying this, not that, and using hand swipes to be able to do those things. So all that's saying is that we gotta have our data prepared to being able to do a lot more with it more efficiently, and, more expansively, not just looking at our internal data, but our external data and AI is the way to accelerate the capabilities that I can do around that data. Okay. So obviously a huge amount of potential there and a and a lot of benefits there. I I guess the challenge then becomes attaining it or building it. And you you mentioned a little bit earlier about the some of the work involved. And for many companies, it might seem like a a large burden on on development teams. And now I know you've done some writing on on, headless and low code search platform. So do you see that why do you see that as a as a game changer for for less technical organizations? Well, so I always looked at use cases as do I need to build something that mimics a very specific journey or customer experience that our, you know, our marketers have really battle tested, that we've learned from our customers, and I need to build kind of a proprietary way of building an experience out. And so if you looked at, you know, simple ways of doing search, sometimes they boxed you into user interfaces or controls or different ways of expressing things. And, you know, that frustrated developers and frustrated the UX and the product managers. And ultimately we delivered subpar experiences with that. So, I first saw headless search, several years ago, in, in a very interesting space. But, you know, it meant that the developers had full control over the user experience, but didn't have to get into the nuances around how to present information to the search engine and how to query for information. They use the search engine for what it was really well tuned to do, and they had control over the experience. So, I had developers working on the experience and I had a search platform being able to handle all the nuances of creating the right queries against what I was looking to accomplish. So that's headless search very useful when you're building customer experience. But when I need to build something internal and I think about, you know, organizations, when you, you know, you think about internal use cases, I have marketers, I have people in operations. I have analysts, I have expert users. I have customer support functions. I have things that I want to build very lightweight tools to do. Maybe they're going to search on three or four terms and that's it. I have others that I want to expose ten, fifteen, twenty terms. They wanna do a more expansive, advanced search against it. Sometimes I wanna have, a UI built into a platform like, like, a a a Teams or a, a portal environment. Other times I want to put it inside Salesforce or another interface. And so I need to rapidly develop simple interfaces that can be placed directly into the tools that people are using and customize that to the workflow that those users are doing. But I can't have developers going and pro doing all that programming. It's just not affordable. And so when I can juggle both use cases and say I can use headless to build a customized experience, I could use low code to rapidly develop lots of, tuned experiences for my employees and for customer support, place them into the workflow that they're using, and I really end up with the best of both worlds. Yeah. No. Absolutely. I think that flexibility of deployment is so important now. You know, even when I think about our own client base, I think, you know, Kaveo for Workplace kind of started as a it was most prominently seen in Internet revamps. But I think as time goes on, we now see clients saying, you know, we have a lot of employees working in Salesforce and ServiceNow. We have them working in, you know, Slack, web based applications, and they're really interested in knowing how they can get how they how they're able to deliver that search experience right where people are actually working. That's right. So I guess we're running a little short of time now, but I'm keen to know, you know, what advice would you offer to leaders to help them come conquer some of the challenges that we've spoken about today? Yeah, I think, you know, when we started this notion of employee experience and responding to how do you make employees happy, productive, successful in hybrid working, how to avoid turnover, and also how to derisk against tribal knowledge. Right? We know people are going to leave and hopefully not the our most important subject matter experts. It means I have to do things defensively to retain knowledge. This notion of employee experience is as big and potentially as important as customer experience was five years ago. We're just getting to it a little bit later and the, importance of it is probably, more compacted because what's happening, post COVID and going into this hybrid work. The problem is nobody owns it. Right? In talking in my interviews, you know, there's different factions who have a stated interest in it, much like customer experience. You know, markers have a a stated interest in it. If you have a product management organization, they have an interest in it. User experience has an interest in it. Technology has an interest in it. But it's really about those groups coming together to come up and say, we're gonna, we're gonna collaborate about finding the right way to plan, build, strategize, deliver experiences for our customers and employees. Every, interviewer told me we can't afford to invest the same amount of dollars and attention and to explore employee experience as customer experience. That's just a reality, but the impact is that still nobody owns it. We still have a lot of, so if you're a CIO, if you're an IT leader out there, I think your job is to draw attention to this gap. Your job is to realize that what you used to do in terms of end user computing that went into the application space in terms of modernizing applications, in terms of modernizing workflows and in terms of, being able to do automation. Now take it up one more step and look at it from the eyes of your employees who are trying to get work done. They want to work from home three days a week. You want to help them make that successful to retain people. We know happy employees are more productive employees. And I think that's essentially my message to IT leaders is, you know, take ownership of this problem and find efficient and productive solutions to deliver to your employees around it. K. Okay. Look. I'm gonna, I'm gonna open the floor to questions in a moment, but I I just wanna just wanna run over a couple of things, first, so to to all the registrants today, we're gonna share with you afterwards this ebook, which kind of gives you a bit of guidance in terms of how to deliver AI powered employee experiences. And what it does is it it gives you examples of where you can actually deploy these experiences. So whether that be, modernizing and personalizing an Internet, unifying self self-service portals. I mean, think about things like case deflection. Right? So if you're able to answer a question for an employee before they have to submit a ticket, then you're lowering that burden on the HR and IT departments. Knowledge in the flow of work, we we we just talked about, you know, how how do you bring that experience on a on a kind of micro level into where people are actually working, integrating it with the tools they're already using to further reduce those friction points. And then how can you use AI powered search to actually identify content gaps? Right? So, you know, if you if you don't have that reporting, then content gaps can continue to exist and continue to to frustrate employees, for a long time before they're actually addressed. And just to kind of give you, you know, an example of some of the Internet some of of how this you know, how the modern digital workplace looks. I mean, like I said, a lot of people come in, with looking for AI powered search to modernize an Internet or a knowledge hub or or a portal. And then, you know, that you're looking at how can we actually provide this experience where people are working in Salesforce ServiceNow. We're continually seeing more, sort of off the shelf Internet platforms. How can you leverage conversations as well? Right? I mean, you've got a lot of employees that are communicating in tools like Slack. And, you know, a a this kind of asynchronous communication is gonna be incredibly important in the future of the workplace. So how do you actually leverage those insights from conversations, bring them into the search interface in order to bring extra context to self-service issues that employees may have, for example? And then like we spoke about, if you have your employees working in web based applications, how do you bring that search experience where they're actually working? Places like SharePoint, Jira, Google Drive, Asana, tools like that. So that's really what we have today. Thank you so much again, Isaac. We're happy to take any questions if you if you wanna leave them in the chat. Otherwise, Olivia, I'll I'll hand it back over to you. Oh, I see one here. So Allison Donald says, can you talk a little about some best practices for measuring the quality of search results? We use the search relevance and content gaps reports available in Conver. But what about evaluating NDCG at ten precision recall? Are manual evaluations of search results, still important necessary? By manual and evaluation, are you talking about kind of a a is this is this useful tag on a on a result? I mean, we we we basically measure relevance using, average click rank, primarily. It gets a little bit more complicated when you're using smart snippet technology, obviously, because then the employee sometimes is getting the answer without having to actually, conduct a click or take any action. So I would say, you know, what we see some of our clients doing is they're measuring this the length of a session on their self-service platform, for example. So if they're seeing employees spending more time on the self-service platform and less ticketing going through to their HR IT department, then they're gonna see they're gonna take that as a signal that their self-service becoming more effective. I'm not sure if that answers your question. I mean, I I could certainly, take some information from you, Alice Allison, and we can we can we can get some more information to you afterwards if if that would help. Go ahead. Yeah. I have two suggestions there. You know, the most simple thing I think is to ask employees, you know, if this is an employee experience or ask customers if this is a customer experience. You know, there's no shortage of noise that you might get from that, but you also get some insights. So when you're dealing with anything that's abstract and you're really trying to understand experience, I really think having a discipline of asking people for feedback is really important. And you'll get some surprising qualitative and quantitative results if you do those surveys well. I think the other part when particularly around employees is, you know, if if I'm given a tool and it doesn't work, chances are I'm not gonna go back and use it over and over again. Right? And so, you know, for any tool that IT puts out, you can start measuring what the repeat usage looks like. And if you see starting to see, you know, people have tried it one, two, or three times and then a float falls off of that, you probably know you have a problem. If they're coming back to it multiple times during the week, they're looking at different sorts of information. They're not doing the same query ten or twenty times because they can't find what they're looking for. So if you're starting to see repeat use against different use cases, then that's probably a good indicator that they're finding value in the platform that you're offering to them. Isaac. Yeah. You're welcome, Allison. So unless there's any more, I think we'll we'll wrap up here. I greatly appreciate, everyone's participation today. And, Isaac, thank you again for your insights. It's incredibly valuable. Thank you. Thank you for joining today, everyone. Thank you. Thank you, Kopail.