Semantic Layer – A Fireside Chat

What I want to get to is our customers being able to have a conversation with their data. They shouldn't need to know where it is, how it's stored, especially with AI, which you use. That's what they want. They just wanna be able to ask a question and move on. That's exactly what the business user wants. Right? To, to ask a question of the data.

But it's not that easy if you, if you, if you want to get that, because there are, there's semantics, you know, involved and there's a, you need a common semantic layer so that we have the same understanding and we say, show me this invoice, or show me the sales order. I think that we are missing that layer. So One of the challenges that we've seen as our customers have attempted to adopt AI is you've got to spend a long time building out a sematic layer, right? And, you know, understanding the business and understanding every part of the business. So the different domains, whether it's finance or marketing or accounting. What we're doing within the intelligence platform is automatically populating that sematic layer. Oh wow.

Because we have actually scanned all of your data assets. We understand how you transact business. You've helped us fill in the glossary, right? The, uh, the role of the data steward in filling in that glossary. So we have a common understanding. What do we mean by customer? What do we mean mean by average lifetime to value?

So do you derive this thing from somewhere? How do you do that? So completely automated right now. Um, but with the ability for the data steward to edit it. If we've gotten something wrong, um, or if they understand the use is slightly different as to how we've interpreted based upon dashboards, they've actually written and delivered to the business and they have the ability to, uh, to edit it. But this allows us then to build this very rich sematic layer so that you can enable conversations with the data. Because ultimately that's where we want to get to.