Detangling Data Chaos: Strategies for Enterprise Leaders to Approach Data Modernization

In this conversation at the 2025 Operations Conference & Exhibition, Jonathan Reeve, Chief Product Officer at BetaNXT sits down with Katie Gibney, Head of Product, Trade & Cost Basis at Vanguard, Bhavesh Khator, Head of Post Trade at DriveWealth, and Val Wotton, Managing Director and General Manager, NSCC, DTC & DTCC Institutional Trade Processing at DTCC to discuss how they are tackling data challenges and how they see open architecture, real-time data availability, and AI-enabled automation as transformative for operations, business intelligence, user experiences, and enterprise growth.

Jonathan Reeve

Good morning, everyone. Thanks for being here. And I'm very excited to introduce, I have an awesome panel here today. I'm looking forward to getting into one of my favorite topics, data. So to my left, I have Katie Gibney, who's head of Trade and Cost Basis of Vanguard. Katie's an avid hiker in her home state of Arizona. Directly to her left, I have Bhavesh Khator, head of post-trade at DriveWealth. He likes to run and recently finished the 50th anniversary of the New York City Marathon. And finally, Val Wotton, head of Equities at DTCC, has actually run three marathons, and in 2026 plans on running the London City Marathon with his son. So thank you to my esteemed panel for being here, and we'll dive into data. Data's a huge opportunity, but also a challenge. We have created more data in the last two years as a planet than we have in the entirety of human history. And so that data is growing, we're all dealing with it, and it's probably safe to say some of the systems that we're dealing with are older than than two years, and they have to keep up with that. And so the topic for today's panel is, what are these firms doing in terms of handling that data? The opportunity, every panel up here is talking about AI and what a great opportunity that is. It all runs on data and data. And data, I think, is very much like water. You need a lot of it to... You need it to live and to survive and to run your business. But if you don't have the right tools, in terms of managing it, then you can also drown. And so many of us are dealing with that challenge. And so I'll pose the first question to Katie here. Every firm is working on this data opportunity. What are you doing in terms of the challenge at Vanguard? What are the top one or two things you're doing to drive this opportunity at Vanguard?

Katie Gibney

Yeah, and I'm happy to kick us off here, so thanks for having us as well, JR. So in 2020, Vanguard embarked on a multi-year transformational journey to take our systems to the cloud, and that's inclusive of our data as well. And it really was driven by a few primary goals, right? So increased scalability. So we have seen over the past decade plus just a significant increase in our brokerage trading volume. So we knew that we needed to ensure that we were prepared to handle even greater volumes in the future, but also for, I would say, prepare for unexpected surges in the market as well. So we think about those surges that we experienced all at the beginning of April. Second would be to improve resiliency, right? Truly raise the bar on our systems availability so that we can meet our client demands and their evolving expectations. And then I would also say for third is just increased and accelerated agility. So cut our time to market and be able to deliver on new innovative offers at a faster rate, which is ultimately going to enhance our client experience as well as be able to put us in a position to where we could drive client alpha.

Now as I think about that modernization for our brokerage trading platform, there were two principles that we anchored to, to help us achieve those objectives. The first one, datamod opens the door, but it's really full stack modernization that's going to get us to our destination. So legacy, on-prem, monolithic, I think those are all terms that everyone in this room has heard within their organization, and that was us in 2020. And so I think what we needed to do is to really transition from that monolithic architecture to a three-tier architecture that was comprised of three distinct layers. So you think about our client experiences, our APIs, and then our data that fuels those experiences. And then two, I would just say, "Hey, design in a way that you can deliver early and often." So we knew that this was going to be a significant undertaking, transitioning from legacy infrastructure to a highly distributed system. And so developing a strategy that would allow us to deliver value incrementally along the way and within our risk tolerance was just a top priority and something that we used to ladder into those broader objectives, JR.

JR

Great. Val, let me turn to you. What's DTC doing in terms of top objectives around data?

Val Wotton

I think I'll just compliment Katie. I think that what Kate laid out is the foundations for every firm around how you modernize your stack, right? Moving from that monolithic, you know, Cobalt kind of challenges that we all face to that distributed model, leveraging cloud more broadly. I think when we take a step back, for us it's all around how do we maximize the value of the data that we have, both from an internal perspective and externally. I think we're very conscious, we're very fortunate to sit on a huge pool of data, but now it's around, how do we leverage that more broadly? From an internal perspective, I think it's all around, you know, really through two or three lenses. One is from a risk and control perspective around resiliency. You know, we've talked a lot about the volatility that we've seen recently. What we embedded through the move to T+1 more broadly across all of our applications is observability tool. So we can actually monitor in real time the performance of all of our applications, and, you know, and we're in a position to make real-time decisions around that, but we want to go beyond that. For me, the next step for us is around how do we use more predictive analytics? How can I monitor the inbound submission behavior of my clients? How can I actually potentially call a client in advance without even knowing that they may have an issue. And so we get onto more of a proactive footing around how we engage with data, and we leverage that across the community, which for the benefit of the industry.

The second bit internally is all around product development. You know, data insights help you develop better products, you know, and for us, those products all around how do we drive further STP automation, minimize risk, minimize fails more broadly. And then that pivots to the external piece, which is how do we provide, make that available for our clients? How do we provide sort of ease of access, the ability for clients to get into data from an access perspective to be able to do their own self-discovery. I think we're all obsessed around creating analytic packages or metrics, but the reality is most clients want to be able to get a hold of that source data and be able to run their own analytics and metrics off the back of that. And I think that's really important. So those kind of areas.

And then the final point I'd probably say is that it's around how do we co-create with data? The reality is the power of data when you put multiple datasets together. So from an equity business perspective, we are very much focused around how do we look at data through the lifecycle of a transaction, from a block and allocation perspective, all the way down to settlement and clearing. Now I have that data, that's great, but we're a global company. We're very much focused on the move to T+1 in Europe and in the U.K. So the reality is how do I partner with other providers, other FMI, other fintechs around creating that data source so then we can look at where there's opportunities to grow efficiency, STP, and automation as well. So very exciting time. We just need to really maximize this opportunity we have with the value of the data that we have at hand.

JR

Bhavesh, let me kick this next question. In terms of working with partners and clients, what are some of the fundamentals you see them working on in terms of their data strategies, and where are the opportunities perhaps to accelerate some of those fundamentals?

Bhavesh Khator

Sure. So drive wealth up. So pretty large institutions, neobanks, fintechs across the board, and their first strategy is to build mobile applications with, you know, API architecture, and with a real focus on speedy delivery to the market. So when it comes to customer experience, they're focusing on making sure that the applications are very simple and easy to integrate with. The onboarding of the users is seamless. Access to data has to be real time, right? Applications are on fingertips. Real-time access to financial markets, being able to execute trades in sub-seconds, as well as being able to look at your portfolio, you know, in real time. And to top that off is 24-by-7 accessibility and availability of the platform, which is pretty key.

And the second thing which they're looking for is building out products efficiently and with speed and launch into the market. It's a pretty competitive space out there. What does get overlooked often, and it's not specific to our partners, but in general is just the data hygiene, right? I mean, we all been talking about it, data swamps, right? In order to fast-track the user experience, sometimes how the data originates, how it gets transformed, where is it consumed, gets lost. And that really is the opportunity as we build the AI. So being able to innovate on a shaky foundation is very, very difficult. So for us to have clear, concise, consistent data, centralized, is keyed for AI. In absence of that, AI models will have errors, automation will create more noise than signal, and your analytics will be more reactive than strategic. So how do we do that? Having a standard data interfaces is key. DriveWealth provides standard APIs with very clear documentation, very good schema definitions, types, and error handling. A lot of the validation upfront is important because once the data gets into your system, cleaning that up is a lot more expensive than not taking it to begin with.

The second is, once you have the data, how do you make sure that it's accessible and propagate into your systems as real time as possible? So event-driven architecture is a big thing we are focusing on. Once the data is available to your system, like centralization is important. And that does not mean just putting the data together in one physical location, but to really integrate your data at different sources, homogenize them, because as we will see, right, the context of the data and the relationship between different domains is really important to power your AI. The last one is data democratization, which means that data is available to all of your internal stakeholders equally in a consistent way. So it cannot be limited to engineers or data scientists, but operations products, they all need to have access to clean and centralized consistent data.

JR

Thank you. Katie, same question. Opportunities to accelerate with partners. Thoughts on that?

Katie

Yeah, so I would say, I mean, a ton resonates with what you just said, Bhavesh, specifically around whether that be the need for data governance or the importance of data lineage, and even just accessibility to data broadly across the enterprise. So I just add a couple opportunities that are in a bit of a different direction, and that's data independency mapping as well as incremental value. So when we think about it from a data independency mapping perspective, right? It's critically important that you under... Like, before you make any changes, that you understand how your ecosystem is connected. And that's from kind of the frontend all the way to the data because, I think, we've probably all experienced that even a slight change can create ripples through your ecosystem and your organization. And then from a incremental value perspective, and I mentioned it just a moment ago, is modernization isn't a light switch, right? And if you try to undertake and modernize the entire stack at one time, it's not only incredibly risky, but it also is going to be incredibly slow as well.

So I think, from a trade platform perspective, one thing that we did is, you know, I would ask you to break it down into three main components. So you have order entry, order validation, and route out to the street. And so as you can imagine, in a legacy infrastructure, those components were so tightly connected that we needed to be really intentional with our design and our approach to kind of untangle that data that was interweaved throughout that monolith. And so I would say one thing that we did was focus on how we were going to do that. And that was through creating abstraction layers to where we could be begin to separate the data differently and start creating those layers that I talked about a little bit before as well. So that's client experience, API, and data. And when we are able to do that and they were able to operate independently from each other, we were able to modernize and operate at different velocities. And I think that was like key for us at Vanguard because we were able to, it gave us stability, it gave us confidence in our architecture, and it also gave us confidence in the data and insights that were flowing now back in between these layers. And so not only were we able to bring new offers to life, into the market, it also positioned us really well to adopt gen AI into our now what will be a robust API ecosystem.

JR

So, Val, I mean at the end of the day, it all comes down to our people, right? I mean this is incredible opportunities, data volumes that we've never seen before. Also blessed with some technologies out there that actually allow us to deal with this amount of data. But it comes down to our teams, we have to get the data teams and the leadership teams aligned around, you know, working on these opportunities. So what are your thoughts around how do we get those teams aligned and what are the sort of philosophies around talent management to sort of, you know, face this new frontier?

Val

That's a really good question. And I think we reflect on this quite a lot over the last 12 to 18 months. I think if you take a step back. Sometimes it's like data's an IT thing, right, and it sits within technology. But the reality is from a business perspective, it's the core to my business. And so we made a very decisive decision around creating data offices aligned to businesses. So we have an equities data office at DTC and also ensuring that that aligns and reports into the business. So it's not a technology function, it's a business function. because fundamentally, it has to drive outcomes from a business perspective and it drives business value, as I said before, whether it's internally or externally. And so I think that has been critical and really just reaffirming the importance of data as a business. And I don't necessarily mean in a sense of trying to create data products to sell in that way. I mean, as a core component around delivering on modernization, delivering on product development, insights, analytics more broadly.

And so then the idea being is then to ensure that we do then engage from an enterprise perspective, and so the data offices, from a business perspective, then engage with where we've created an enterprise data office to be able to ensure, again, collectively, we're engaged across the entire enterprise, alongside an AI council as well. And so I think that structure in place then helped leadership get a real understanding of what are we delivering from a data perspective. And I think most importantly, being able to help when it comes down to prioritization, funding discussions, and really then what we've also done is look to embed within our overall corporate objectives, data goals as well. So the idea really is from top of house all the way down, it's embedded.

I think education is then really important. And last year, we sort of kicked off an external provider, an eight-week program for the senior management team, really engaged, you know, educating us all on AI and all facets of it as well. So fundamentally, you know, that that whole piece of upskilling ourselves as well so we can make well informed decisions around how we leverage that within our own enterprise or within products within the business too. So for us, I think, from a leadership is really around connecting from the most junior to the most senior, being aligned as an overall organization, and then helping to make those decisions around priority as you may forward.

JR

Bhavesh, same question in terms of aligning teams at DriveWealth. What are your thoughts?

Bhavesh

I think Val pretty much answered that, right? Yes. I think we are evolving in a way where data is no longer an outcome of a business process, right, but it's becoming a strategic asset. And I think the leadership which really embraces the fact that data is now a product, right, not an IT concern, right? That naturally aligns with the data organization. On the flip side, the data organization also need to use the data as with the context of solving the business problems, right? How we modernize our data infrastructure, right, using the problems at hand, right? So that goes a long way. So instead of, you know, the goal would be modernizing ETL pipelines to completely reverses look, you know, in order to improve our, you know, business operations or, you know, cost basis or whatever, right? We need to integrate to a modern architecture, and that's how we solve business problems.

From a talent side, I think it's pretty evident that in the next few years, the talent we need will completely change with the advent of AI. So I think the first thing is to really understand what skills we will need going into the future to really use AI. And one thing which is clear that people who have technical specialization will give it to people who have a lot of context of the business, right? So instead of building code or writing code, the focus will be on people who can ask what data can do for you, right? Can you ask the right questions? And that obviously needs to go into the hiring process. When you are recruiting for talent, you need to, along with the core skills, you also need to make sure that there is adaptability, right? People can learn new skills faster, right? Critical thinking is required more than ever before. Within the organization, as Val mentioned, right? Making sure that data literacy is a key priority across the organization. Re-skilling people with data fundamentals, including what is new with the AI, and the tooling, I think, you know, how to use tools and so on is really important. And last but the not least is also how you choose your platforms, your technology providers is really key because that can also help minimize the work on your side and reduce that skill gap.

JR

Music to my ears. Katie, what are your thoughts about, you know, rallying teams around modernization initiatives? What's going on at Vanguard?

Katie

Yeah, and so Val hit on a lot of it just from a kind of leadership perspective as well, but it sounds so simplistic, but it is really having a shared vision across the enterprise and anchoring to the “why”. So Val mentioned, hey, it's not just an IT thing, or it's not just about the software, right? Modernization is a business imperative. And for Vanguard, our “why” is, we are uniquely client-centered, and our trading platform is the primary means in order for us to distribute our products to our retail direct clients, reinforcing Vanguard's investment philosophy. So I think just the alignment on modernizing our ecosystem just reinforces the importance that we as an organization place on driving our mission forward and giving our clients the best chance for investment success. I will also say that it's not just vision alone that's going to get you there. Structure and discipline are really important. And that's when a strong product operating model comes into play.

So we think about, for us, it's going to drive consistent prioritization, it is going to empower teams to make decision, and these are teams that are agile, these are teams that are data-driven, and these are teams that are aligned to our client needs. And so giving that power and like when you do that right, you are bringing together a culture that is technology, business, data, and design to feel ownership and to solution together. And I think it's just when those teams have that ownership and solution together, the momentum just continues to build.

JR

Exciting times for sure. Val, DTC has long been at the center of the ecosystem that we all work with. So what are you thinking about in terms of the DTC roadmap, building a more open connected ecosystem for all of us to operate in this world of exploding content?

Val

I think, you said it is exciting, right? I think as we go through our modernization journey, as we've just laid out here, it's around how we can deliver value to the industry. And I think there's probably two or three key examples. The first one is by us moving to a position where we can stream real-time data, it means that we can support 24-by-5. And we made the announcement, you know, for Q2 2026, the NSCC will be in a position to support 24-by-5, we'll operate from Sunday at 8:00 PM through to Friday at 8:00 PM. That we could not do on legacy technology. And so that's what for me excites me is I'm delivering industry value, very similarly on the DTC side, right? You know, the demand for us to be able to support partials, you know, to auto-partial within DTC has been there for many, many years. Again, as part of the modernization agenda, we'll be able to deliver on those things, and all the tooling that comes with it as well. The collateral monitors that we have built and we need to continue to improve upon all around facilitating the industry and clients be able to make better decisions.

So I think we are a very exciting point where from a product and business perspective, the ideas that we have and we want to be able to deliver, we can look at trying to accelerate where possible, but fundamentally, embed it into our modernization journey. I think that's where you get the sweet spot is for me it's around delivering business value underpinned by modernization, core applications but really driven from a data perspective. 'Cause it all comes back to the data in my mind.

JR

Bhavesh, you work a lot with partners. DriveWealth has a brokerage as a service platform. You're dealing with that integration all the time. How do you think about how to facilitate that partnership integration and accelerating that with some of the people that you work with?

Bhavesh

Yeah, great question. So DriveWealth deals with partners across large institutions, new brokers, fintechs. And even though we have a ready-to-use brokerage service platform, we really use a white-glove approach to integrate our partners. We are not just a technology provider but their partner in execution. And that really starts with right at the inception with integration, we have key personal, solution engineers, and project managers who really track the progress of each integration, even though most of the work is on the partner side. I think we have standard templates, which really based on the part partner persona, give them a very good indication of what should be their core feature set versus a good to have and something which they can launch in the future. That really defines, brings clarity of the scope, and helps with the speedier execution.

The next one is very easy to use, very well-documented APIs. They're publicly available with integration guides. So developers can really use that and kick off the development without a lot of the handshakes. On top of that we also have build integration templates. So, for example, if you want to launch a feature, book a trade, there's a standard recipes which says, "Okay, these are the APIs you need to integrate to be able to book a trade." It starts with, you know, creating a session, creating a customer, creating a trading account, place an order using an order API, and get the order back using an order API. And that enhances developer experience and obviously helps speed execution of the integration. We also minimize how much data we need to collect. So we focus on the regulatory and compliance, but need or data is used only on a need by need basis. And more often than we use what partners already have so they don't have to reach out to the customer again.

And on the flip side, partners have access to the data via APIs, via events, via, you know, a shared cloud infrastructure, dashboards, batch files, various ways they can access data so they can build their data platform their own way. And last but not least, as partners can integrate their standard integration testing, which partner has to go through before their launch. This whole thing is a singular focus on making sure that partner is seamlessly integrated with us, and their focus is less on the integration, but more on their customer experience, their product launch.

JR

Well, it really is a new frontier in terms of data opportunities and challenges. You know, the growth of data as I started with, we've added more than more data in the last two years than we added in all of humankind. And so you can imagine what that's going to look like in the next five years, the next 10 years. So our systems obviously have to deal with that, and it looks like people are starting to think with it. Think about it. I know, certainly at BetaNXT, we're building a lot of solutions to deal with those challenges.

That's the end of this session. I'd like to thank very much my esteemed panel, and thank you all for your kind attention, and wish you all a good day.