Metadata as a Strategic Asset | Terrence Hedin, LSEG

David Sweenor 0:04 All right, hello, welcome. Hello. Welcome to the Gartner data and analytics 2026 conference in Orlando, Florida. I’m here with Terrence edin. He is data and metadata platform director at London stock exchange group. I hear you giving a talk tomorrow.

Terrence Hedin 0:23 That’s right, yep. So I’ll be speaking with Phil. Phil Dutton from solidatus. What we have lined up is kind of a bit of a lightning talk. So we have a lot to cover in a short period of time, but it’s really going to be an introduction into how we’re approaching managing our data properly at lseg in many different ways, and the business features it enables, including the AI readiness side. So we’ll be talking about how lineage provides value, how we use solid datas to create that lineage and to get the visuals on it, and that visual piece, the real how important that is. I’ll provide a focus on our Data Trust program and the four elements of trust and how that relates to everything that we’re doing that’ll be highlighted throughout my talk. The next big piece I’ll cover will be our business contact context piece. So how we bring our business metadata, our technical metadata, our semantic layers, into a knowledge graph so we can build that true business context that provides not only human benefit, but machine benefit as well. And on that machine side comes, of course, AI, and one of the aspects to that is how we deliver that entire business context and Knowledge Graph as a machine readable feed aligned with open standards across the industry. Last piece of that will be our approach to treating metadata as a strategic asset, and so that means bringing true business value from this thing we call metadata, treating that metadata as a data set, publishing it as data products. Yes, real data products, not for data mesh anymore, not fake ones, not fake but even making that commercially available as a product to our customers, so both internal and external customers. All of this wrapped around how we’ve taken, traditionally, metadata being for governance, not just for governance anymore, providing real value add on top of it, and building off that business context. And some of the things that we get from solid data around automation, the visualization and our ability to really, truly understand what our data is, and then Larry on top of that, what it means, and then putting real value on top of that,

David Sweenor 2:26 that’s amazing. So Terrence, I just have a quick question for you. So you know, prior to this age of AI, we’re in lineage and metadata were sort of, I don’t know second class citizen is the right way to phrase it, but they’ve increased in importance and prominence. I think. Why is lineage important for AI and metadata for that matter?

Terrence Hedin 2:49 Okay? I love that. You called it a second class citizen, okay? Because it has evolved. It is a first class citizen. Yep. So we in every business requirement spec includes lineage at an element level, every tech spec includes how do you produce that lineage? All of the real value is encapsulated in releasing products where we’ve established that lineage. And the reason why is because if we don’t understand what that data is, it’s very difficult for us to understand how we can use it, how we should use it, what value it can provide, what new products it can influence. It used to be the simpler cases where, well, what if a customer calls you up and asks you where a piece of data came from? Right? You have to be able to answer that, right, and

David Sweenor 3:32 that was for regulatory reporting mostly. Or my reports messed up all the above,

Terrence Hedin 3:36 all of those. Yeah, those were kind of the original, as you put it, second class citizen, right? It’s always a first class citizen, but the way that we approach it, that’s kind of really the evolution. And so, yeah, I mean, it really is that foundational piece for us to be able to understand what our data is, and that hugely improves and accelerates our ability to produce products, and accelerates our ability to there, reduces the time to market so we can get them out the door, and it also gives us that confidence that we know what we’re doing. Just a quick story, and the same with your second class citizen. It used to be the case where we would ask a data product owner, you know, at what point are you capturing lineage for that? And the answer might have been, you’ll have to ask someone who knows about lineage, right to which we respond? Well, do you need to know where the data came from? The answer is absolutely, and I can tell you, or I can get you a spreadsheet that shows it, whereas today, it’s part of a completely automated, not 100% automated, I don’t want to paint the wrong picture, but a very automated business process. It’s complex, it’s not easy, but it’s accurate, it’s reliable, it’s consistent, and that’s what we need when we build our lineage, to put everything else on top of it. That’s amazing. And maybe

David Sweenor 4:47 one last question, and if you can’t answer this to your ratio roll, that’s fine, but, you know, solid data is just announcing, you know, their new AI lineage assistant. How has that changed? Changing the nature of what you do or how you approach it. You know, a lot of automation and smarts built into that. I’m just curious.

Terrence Hedin 5:07 So that’s really key, and it’s one of the reasons where solid attest has played, or why solid attest has played a big role to date, is the older days of drawing pictures on napkins, spreadsheets going on, word of mouth, protected knowledge about where it, data came from, what happened to it on its way out the door. Part of this consistent, accurate, reliable story is it has to fit into the processes that we put in place right. There’s still plenty of manual activity there. The more that we can bring in agents and AI to help resolve some of that, in particular, driven off the business context that we’re creating already on top of what we know, there’s huge opportunity to continue to continue to automate that. And it’s it’s not cheap today, so any ability that we have to simplify and make it more efficient is very attractive.

David Sweenor 5:54 All right. Well, Karen said eating your talk is tomorrow. At what time it’s at 335 335 in theater, three, three, so make sure you check that out at the Gardner data and analytics in Florida. Thanks. Thanks for the time. Terence, thanks for having me. Cheers.