Hi everyone. I'm Harshita Doe from Apex FinTech Solutions former with Technologies, and I'm thrilled to be here at conference 42, data Script 2025. Today we are diving into how embedded analytics and data scripts are transforming enterprise knowledge systems, turning what used to be static repositories into dynamic intelligent ecosystems. Let's start with the problem. More than 50% of the enterprise knowledge systems fail to deliver real value. We are talking about massive organizations, a hundred K plus employees across hundreds of countries, spending millions on tools that people don't actually use. Our journey to fix this took 24 months, two years of transformation to make knowledge accessible, contextual, and measurable. Why do these systems fail? First, static architectures legacy tools can't keep up with how modern themes actually work. Second, disconnected content information lives in silos, making it nearly impossible to find what you need. And third, poor user experience. If finding information feels harder than asking someone on Slack, people won't self serve, and that's the reality. So how did we turn things around? We approach this like a data-driven product transformation, not just an IT project. We embedded analytics to understand user behavior, mind search patterns, to uncover intent, built real-time feedback loops, and enhance SharePoint with custom JavaScript microservices. In other words, we stopped guessing and started listening to the data. JavaScript was our engine for transformation. We built dynamic UI components that adapt in real time to user behavior, context aware filtering made sure people saw only what was relevant to them, and natural language search allowed users to ask questions the way used humans too, not the way research engines expect. That's how we brought intelligence and empathy into enterprise systems. Here's a quick look under the hood. We use SharePoint online as a secure foundation, and on top of that, we layered custom JavaScript, microservices handling analytics, search and personalization. The analytics engine processed live data for insights, while the UI delivered a responsive, tailored experience. It's modular, scalable, and measurable. The results speak for themselves. We got information retrieval times by 33%. Self-service resolution jumped 22 points from 62% to 84%, and onboarding time also dropped by 21% in short. Employees found answers faster. Got up to speed quicker, and less on support. From a business per standpoint, the impact was just as powerful. We saved $3.7 million annually through efficiency gains support. Tickets also dropped by 35% because users were solving problems themselves. Analytics didn't just improve experience. It drove measurable. ROI. This transformation also changed how we governed content. Our dashboards track engagement. What's working? Watch it note analyzed where users abandoned searches to fix broken experiences, and we detected emerging needs before users even request new content. It's governance powered by insights, not instinct. At the core, three engines drive the system intelligent search powered by NLPA personalization engine tuned to behavior and roles and real time analytics dashboards that kept everything transparent together. They formed a living learning system that evolved with users, not despite them. If I had to summarize what really made this work, it's these four principles. First, modularity build independent service pieces that grow without breaking each other. Analytics first, that is measure everything from day one. Third, scalability, assume success and design for it. And fourth and finally user-centric development because tools fail when people aren't part of the build process. Here are a few lessons that stuck with us. Start with user behavior. Don't just assume, invest in microservices. You'll unlock flexibility without even replatforming and make data visible. Real time dashboards create accountability and momentum. These lessons are universal, no matter your tech stack. If you're wondering how to get started, here's the roadmap that we used. Assess where you are today. Lay the foundation with analytics. Add intelligence with JavaScript powered features. Optimize based on feedback, then scale across the organization. It's a repeatable, proven approach, not just a one-off success story. Let's wrap it up. First, embed, analytics and everywhere. Second, leverage JavaScript to make systems adaptive and intelligent. Third, focus on measurable outcomes that prove ROI. And finally, build modular, scalable architectures that evolve with your business, and that's how we transform knowledge into action. Thank you so much for joining me today. I hope this session sparks ideas on how you can make your own knowledge systems more intelligent and human centered. You can find me on call 42 or LinkedIn, and I'd love to continue the conversation. Thanks all. Bye.