In this episode, Neelesh Prabhu, Managing Director of Architecture and Enterprise Services in the Information Technology Group at DTCC, talks about modernizing legacy technologies, leveraging the relationship between business and IT, the future of the financial industry, and much more.
In this episode, Neelesh Prabhu, Managing Director of Architecture and Enterprise Services in the Information Technology Group at DTCC, talks about modernizing legacy technologies, leveraging the relationship between business and IT, the future of the financial industry, and much more.
--------
How you approach data will define what’s possible for your organization. Data engineers, data scientists, application developers, and a host of other data professionals who depend on the Snowflake Data Cloud continue to thrive thanks to a decade of technology breakthroughs. But that journey is only the beginning.
Attend Snowflake Summit 2023 in Las Vegas June 26-29 to learn how to access, build, and monetize data, tools, models, and applications in ways that were previously unimaginable. Enable seamless alignment and collaboration across these crucial functions in the Data Cloud to transform nearly every aspect of your organization.
Learn more and register at www.snowflake.com/summit
[00:00:00]
Steve Hamm: welcome, Neelesh. It's, it's great to have you on the podcast today.
Neelesh Prabhu: Thank you Steve, and I am thrilled to be on the podcast with you.
Steve Hamm: Great, great. Now DTCC plays an incredibly important role in the global financial markets, yet I'm sure that plenty of people listening to the podcast today aren't familiar with the company. So tell us about it. Who owns the company? What is its history and what services does it provide today?
Neelesh Prabhu: Steve, uh, DTCC stands for Depository Trust and Clearing Corporation. We are a. Year post trade infrastructure firm for global markets, providing stability, certainty, and reliability for the financial services industry. And we support our clients by mitigating risk, increasing transparency, and driving efficiencies.
But we've got a very interesting [00:01:00] history. We were created by the financial industry to solve a paper. 50 years ago, if you bought a stock in a public company, usually it was a stock exchange. You sent in a check and received a paper certificate to reflect your ownership of shares. So in 1973, DTCC was created by the financial industry to put all of those certificates in a world and create a computerized central.
Today we are the world's largest, uh, depository of active US issues, and we clear and settle virtually all stock and bond trades in the US markets, and, uh, that that implies around more than 200 million transactions a day. Across 50 different markets and trading venue. Since then, we've also expanded our [00:02:00] offerings to other financial products, including the global derivative markets. In 2021, we processed over 2.4 quadrillion dollars of security transactions. And um, just so that your listeners understand, I am not making up the term quadrillion. It's one followed up by 15 zeros or a thousand trillion, and that that reflects our scale.
Steve Hamm: No, that's amazing.
Neelesh Prabhu: We're also, um, Steve, we are also regulated in almost every market we operate in.
And, uh, we have oversight by 23 different regulators across the globe. And the interesting thing about our culture is why we are focused on, uh, stability, security, and resiliency. We see the see innovation as one of our strategic pillars. So we really built a culture where innovation is encouraged and celebrated.
Steve Hamm: Yeah. No, that's a great and thorough answer. Thank you so much. Hey, I, I did ask [00:03:00] who owns the company? Is it owned by the, by the people, by the financial services companies, or is it separate?
Neelesh Prabhu: No, it is owned by the, uh, financial services industry and, uh, we have over, uh, 275 shareholders who comprise of various players in the financial
Steve Hamm: Okay, I got it. Got it. And you also described it as this, the, the core service is a central ledger. So are you using blockchain or are you planning on using blockchain in the future for this, for this capability?
Neelesh Prabhu: Yeah, the, uh, as a part of our innovation strategy, Uh, Steve, uh, we have actively looked at blockchains and the ability to leverage that technology in order to further our services. And if you go onto our website, uh, you should look up Project Ion where we have leveraged the blockchain as a very interesting capability to clear and settle, uh, security transactions, uh, here in [00:04:00] the us.
Steve Hamm: Yeah. No, that's really great. That's good. A fantastic overview. So let's drill down to one individual. You, what's your role in the company and what's your strategy for modernizing the way DTCC manages data?
Neelesh Prabhu: so I have global responsibility for both dtcc c's technology, architecture, as well as the delivery of. Enterprise technology services for dtcc, I lead a team of over 1,500 technologists who architect D TCCs applications, data, and platforms, as well as build and deliver foundational capabilities like data that enable both agility and resiliency for all of our businesses. We have embarked on a multi-year strategy, uh, that will modernize, um, not only our, uh, uh, [00:05:00] technology, but also key aspects of our business, including enhancing, uh, customer experience, while managing production stability and resiliency. Data modernization is a key aspect of that overall modernization.
Steve Hamm: Right, right.
Neelesh Prabhu: We are using, uh, that to improve our client experience, uh, providing timely and seamless access to data, both for internal and external clients.
We are also focused on breaking data silos that exist within our enterprise and enhancing data resiliency. Uh, like any large, um, financial services firm, uh, Steve, we run a number of legacy monolithic platforms, and our goal with our data strategy is to enable innovation and agility for our businesses while, uh, these legacy systems are being modernized.
Steve Hamm: Yeah, yeah. No, this is just incredible. I mean, it just, it's just sinking in with me. I [00:06:00] mean, your organization must deal with the most data of any organization in the world and also the most transactions. I mean, you basically are, are putting a real stress test on any technologies that you use, so you have to have like the best correct.
Neelesh Prabhu: Yeah, that that is, that is a very Right. Uh, uh, Steve and we, we work, given our unique use cases, we work very closely with our partners to ensure that. We are, we are feeding in our requirements, which in many places push the boundaries like you mentioned, so that they can build it into their applications.
Steve Hamm: Yeah, yeah, yeah. Now, uh, I was looking at a little bit about your, your career history, which is really very strong. So you've had a series of important data management jobs in financial services. Before dtcc, you were at TD Bank. Capital one C I T T I A a cro, how have your experiences at [00:07:00] those companies and organizations prepared you for the challenges that you face today?
Neelesh Prabhu: I've been indeed fortunate to get a bird side view, uh, of technology infrastructure for financial services companies as well as businesses, uh, around the, around our industry. probably draw, maybe the best way to, uh, describe it is there are certain teams which are common, right?
So firstly, across our in. We are increasingly digitizing, uh, our business and critical services are being now delivered digitally, resulting in ever increasing importance of data. When I started my career many years ago, um, uh, I heard the, the phrase financial businesses are really data businesses, and it could be more true.
Steve Hamm: Yeah.
Neelesh Prabhu: Also, as I, I just referred to most financial services, [00:08:00] uh, companies have large investments in legacy technologies, which lock their data. In, uh, product systems. So modernization of these legacy technologies is a great opportunity, um, not only to modernize these systems, but also modernize our data architecture.
And this requires firms, uh, to act and behave differently, uh, than they have done in the past. And maybe there are a couple of themes I can touch on. Firstly, It is important that firms not only focus on data technology, but also focus on building, um, a another data muscle, which is investing in people, culture and capabilities to exploit the data they have.
And building such a culture, uh, is a hallmark of firms, uh, I have seen who are really successful at driving, um, uh, data programs and [00:09:00] exploiting value.
Steve Hamm: Right.
Neelesh Prabhu: The second thing I would point out, Steve, is that, um, uh, building data capabilities, um, is also a very special, um, uh, art, which, uh, successful firms have. Um, uh, have figured it out.
And the whole idea is to not, um, uh, build data capabilities in a big bank. Do it incrementally and do it with clear line of sight into, uh, how these capabilities will be consumed by the business and how they will generate, uh, business value.
Steve Hamm: Yeah, that's interesting. You know, it occurs to me, you know, you've had a, a good, long career, uh, several, you know, big kind of famous companies you, you work for, and I imagine over the, over the years you've kind of gone through one transformation of technology after another. Cuz these things go into waves and they.
As far as I know, they never stop, [00:10:00] but the, the latest wave, of course is the cloud. So I want to drill back on, on DTCC again. When did the organization start migrating applications and data to the cloud and when and why?
Neelesh Prabhu: Yeah, Steve, we, the, we started down this journey, um, in 2012. So, uh, we, we were, uh, relatively, uh, uh, early to the game and our desire to leverage the cloud, uh, has been, uh, to drive two main benefits. Firstly, leverage tremendous innovation that occurs on the. And secondly, to enable the scale and the resiliency that the cloud brings for our businesses, we leverage the cloud at DTCC under two broad use cases.
Uh, the first one, uh, is, uh, the, our desire to exploit the public cloud, uh, to buy into world class functionality. [00:11:00] And gain cost efficiencies by leveraging, uh, software as a service primarily for our corporate functions and some IT capabilities. So for functions like finance, HR, and collaboration capabilities, we leverage the cloud very heavily.
And the second use case, uh, we've been involved in is, um, leveraging the public cloud. To host, uh, our business applications and given our resiliency requirements, given our unique requirements in the scale we handle, uh, this has been a very graduated journey and we are actively working with our cloud partners to ensure that their services can meet the needs of our applications.
Steve Hamm: Yeah. Yeah. Now, in terms of this migration, the, the maturity of it, I mean, do you still have a lot of stuff, lot of data on premises, or is it mostly in the cloud now?
Neelesh Prabhu: have been on an active journey of migrating, uh, [00:12:00] our data, uh, onto the cloud. So, uh, the vast majority of our data, especially data which is used for, uh, non-transactional purposes and data exploration purposes, uh, we have been migrating that to the cloud at a very rapid.
Steve Hamm: Right, right, Right. Now, DTCC began using Snowflake's Data Cloud in 2019. What were the initial uses and what results have the technologies brought?
Neelesh Prabhu: Yeah.
Steve Hamm: Uh,
Neelesh Prabhu: So we started, um, on this journey, uh, Steven, the 20 18 20 19, um, timeframe, and the prime driver for us was that the on-prem. Data infrastructure we were using was, uh, at end of life. And so we had a major decision to make, whether to current, uh, to continue down that path or to, uh, uh, take a path which led us to the cloud.
Our current model had two [00:13:00] major drawbacks. Firstly, um, scale and expansion. Meant, um, major time to market issues for us. Uh, we had to go through a lengthy procurement process. There was the racking and stacking of the infrastructure in our data centers, and finally, uh, installing software and configuring it, all of which would take months.
The cloud model was very appealing because, uh, that would significantly reduce the time to market and make, um, that process seamless.
Steve Hamm: Okay.
Neelesh Prabhu: um, as an IT organization we had to invest in specialized teams to manage and run this data warehouse, um, uh, software and hardware, which essentially meant, uh, additional costs, but also complexity.
So the cloud, uh, model was a relatively easy, uh, decision for us. So we did, um, as we migrated, uh, to the, to [00:14:00] this model, we achieved significant benefits. But the biggest benefit, uh, which, uh, we are now realizing is one around driving innovation in our businesses by extracting data from these legacy environments that are ever to change to a cloud environment that is designed to have on compute demand and storage power.
Along with modern IT tools to better visualize, analyze, and distribute that information. We've been able to sharpen our focus on customer needs and experience and that has been a, a indeed, a game changer for.
Steve Hamm: Yeah. Yeah. And so you have, you know, your basic services that you provide, and then you are, then you're, basically, the lot of the innovation is in these additional new data analytic applications for your customers, right? And that's what you're doing with Snowflake.
Neelesh Prabhu: Yeah, we, we are leveraging Snowflake, um, certainly to provide, [00:15:00] um, additive services, uh, for our customers, but we are also leveraging it internally. To democratize data and create a culture of innovation internally.
Steve Hamm: Yeah, I see. That's great. So in late September, DTCC and Snowflake announced a significant expansion of the relationship to transform how data is accessed, shared, and leveraged across a number of your services.
So please give us some color on that. What are you doing and why?
Neelesh Prabhu: Yeah. Um, Steve, this is something which has, um, been a very exciting, uh, development for us. Uh, as I mentioned, we already leveraged Snowflake, um, as a data foundation across our businesses. In, um, market utility repository and data services, we believe that we can leverage, um, other core capabilities that exist in the Snowflake platform to position us [00:16:00] to solve key industry and client needs.
One area which is of great interest to us is data sharing. The market is evolving from, uh, you know, sftp FTP messaging that required downloading of data between players in the data supply chain to data sharing models where we can leverage common cloud-based environments. And then multiple people parties can view and interact with the same.
This obviously has, uh, benefits for us. Um, in terms of efficiencies in it. Budgets, uh, we won't clog up, uh, networks. We, uh, won't need as much infrastructure to distribute the data, but it also has significant business benefits under this model. We can, once a client elects to license a product, we can onboard them faster.
In an SFTP world, uh, the provision process can. Four, three to four [00:17:00] weeks and we provision access on weekends. You know, you miss a weekend and you lose weeks of subscription revenue.
Steve Hamm: Right,
Neelesh Prabhu: um, it also improves our ability to develop, uh, what we call intra day products, which is the next phase of our product life cycle.
And these products are more valuable and they carry a higher license than daily products.
Steve Hamm: right,
Neelesh Prabhu: And finally, uh, you know, we've got a very strong focus on resilience. Uh, this data sharing model, um, makes our delivery process more resilient. You have less ops in data delivery, which results in less points of failure.
Steve Hamm: Yeah.
Neelesh Prabhu: We're, we are also, uh, seed. We're also looking at a couple of other areas, uh, like um, running, uh, data bound applications, uh, especially in our risk area, and with the impending move of the US capital markets from. T plus two to T plus one. Um, [00:18:00] we will be able to improve our risk management posture by running intra day analytics and also leveraging the data science capabilities on the Snowflake platform to, to, uh, crunch AI models to some of that massive data, uh, uh, which you and I spoke about earlier.
Steve Hamm: Yeah. Yeah. And Indate, just so I, I'm clear, that means that people could do real time analytics on their data? Correct.
Neelesh Prabhu: Yes. Um, near, near realtime analytics on, on their data. Yes.
Steve Hamm: Okay, good, good. Now I wanted to drill down a one on one aspect of your services. You have the, these data services called Kinetics for a, a variety of different kinds of securities or, you know, fixed.
Income or equities, things like that, help us understand, you know, kind of what is the kinetics portfolio, how does the Snowflake technology enable that to be, you know, more effective and for you and for your clients.[00:19:00]
Neelesh Prabhu: So Data Kinetics is, um, a product offering that provides insights into market liquidity and sentiment for our customers. And we source data from D TCCs Clearing Asset Servicing and Trade Repository Utilities across all major asset classes. And, uh, Steve, given our central role in the global capital markets, We have a unique vantage point because we can see almost all transaction activity in a number of asset classes.
And um, it is this data or this transaction activity that has been anonymized and aggregated. That forms the basis of the kind products. Snowflake has been a great enabler in developing this product, right? It brings us a number of benefits, uh, including lower cost, faster revenue, and time to market, and [00:20:00] improving our resilience along with better entitlement and control and, uh, enabling us to innovate. The fact that we are capturing increasing amounts of, uh, DTCC trade activity in Snowflake. Will make it easier to develop new data products. And the, the really interesting, um, shift we are seeing is the interaction between the business and it, because the cost of this, uh, of developing these new products is shifting from what was broadly it cost to, um, the risk on the business side to make sure that they are making the right decisions in terms of, uh, product priorities.
We are also seeing, um, that um, we are able to reduce, uh, the compute time for workloads. For example, uh, the equities, kinetic products, which used to take three hours to process now takes six minutes. [00:21:00] And, uh, uh, the fact that we can get data faster to our. Has immense value. For example, a, a hedge fund manager in London will have our data in time to run their sentiment models, but the markets, uh, are ready to run in Europe, uh, which, uh, puts them at a significant advantage.
Steve Hamm: Yeah, I think that's a great example, but I, I wonder can kind of explain it a little bit more. I mean, you know, a lot of our podcast listeners are not in the financial services business and certainly not in the kind of data, you know, intense, uh, aspect of it that you've been describing. Give us an example of how a client uses the kinetics package.
Uh, you know, how, how do they use it to optimize or improve their business? What, what are the advantages of timeliness that you were able to offer them? Give us a scenario there.
Neelesh Prabhu: Steve, let me take an example, uh, of our equities, uh, [00:22:00] kinetics product, and in this product we aggregate data. Through the sell side, clearing cor, so all sell side firms, brokers, um, come in every day and send all that trade data to nscc, which is the sell side, clearing corp to, uh, clear and settle their trades.
We then aggregate that data to a stock level and after it's been anonymized, We can then cut up that data by different trade sites. So you can see the data from a by perspective, sell perspective, sell short and sell short exempt. The, the consumers of this data are, uh, typically, um, quantitative investors who are looking at a lot of data points to identify strategies in order to turn the trading profit.
And in this age of big data and artificial inte. They run some fairly complex data models and, uh, we are a [00:23:00] part of that big jigsaw puzzle with they're working with. And the value CB bring, um, to the equation is that we can provide a consolidated view of US equity markets, and in absence of our data set, they would have to go to 27 different places.
To, uh, gather this data. Another great example of how this data set can get used is for sentiment analysis, and you can do that by looking at, um, um, the, uh, charts which are, um, which have been placed on different stocks. And this can give you a sense for what the market feels about these different stocks.
Steve Hamm: Yeah, I imagine there's been a lot of that done with meta recently, right?
Neelesh Prabhu: Yeah, absolutely. Absolutely.
Steve Hamm: But that, that, yeah, that doesn't take a lot of, uh, subtlety to figure out what the sentiment is there, but there are others where I imagine it's. You [00:24:00] know, there's obviously there are people who are, you know, buying along and other peoples who, who, who are selling short or, or whatever.
And you know, there, there are differences of opinions and there a lot of them must be very subtle and it's in the data that you can figure out what the best strategy is for an individual investor. So
Neelesh Prabhu: Exactly. And Steve, another, another example would be just liquidity, right? Which can tell you a lot about, um, uh, uh, uh, what is happening in the markets. Um, our data set lets you. Look at where trading is concentrated for individual securities. How many different players, uh, across peer groups are participating in that trading?
And that is pretty powerful.
Steve Hamm: Yeah. Yeah. That's interesting. So I, I, I realize as you're talking that. Because you aggregate and anonymize your clients are, are able to see the entire, you know, whatever cut of the data they want. Not just their trades, but everybody else's [00:25:00] trades. And so that they do get this whole market view and that is really, I think, extremely valuable for them.
So yeah, thanks for explaining that. Um, you. The, the kinds of things you've been describing are really extremely powerful, uh, uses or examples of data management, data analytics. Where do you see these things going over the next, you know, four or five years? Not just for financial services, but kind of the world of business?
Neelesh Prabhu: Yeah, this, um, Steve, this is part of an overall broader trend where firms are coming to terms with the value of their data and then they're looking at the cloud to enable. Development of new products and the ways in which they can join their operations with customers. Financial services in particular is evolving from a transaction or a payment business model.
[00:26:00] One of providing insights and analytics back to their customers. You see this in your personal life. Uh, if you think about your online banking services, at one point of time they were a way to make payments, But today your bank, uh, can give you, uh, important indications about your behavior. It can track your spending habits.
Uh, you can get personalized, uh, uh, or uh, alerts on fraud. And it can also show you ways to. And within the capital markets, at the institutional level, the same, um, concepts are, are, are, are taking place. We are evolving from, uh, being a, a transaction partner for payments settlement asset servicing to an information partner at through all stages of the investment process.
Steve Hamm: Right.
Neelesh Prabhu: And, uh, uh, banks in the institutional space in general are going to extend their [00:27:00] universe from, um, transaction processing to other areas like compliance, regulatory, deporting, and risk management. And Steve, this is going to blur the lines between the different players, right? Collaboration and data sharing.
Are going to be a part of this trend, and specifically, um, in the financial services industry in particular, but broadly, we are all getting more comfortable in using cloud services, storing data in the cloud. And this is going to create new ecosystems where data can be shared between market participants.
Steve Hamm: Yeah. You know, it's just amazing to think back. I mean, how on earth did businesses operate 20 years ago? I mean, you know, it was almost as if they didn't know what was really going on inside their, you know, within their operations or with their partners, their, their downstream distribution, all that kind of stuff.
I mean, obviously, [00:28:00] For the time it was satisfactory, or it was, or it was. It was, you know, what, what was available, but it's just, it's just amazing to see this revolution happen kind of in real time all around us and with people like you driving it. So this is really a cool conversation. Um, you know, we typically end with, um, a lighter, more personal question, and I hear that you're something of a history buff.
How do you compare what's happening today with data and technology to other waves of innovation from the past?
Neelesh Prabhu: Steve there. Uh, there are a lot of parallels, right? And, uh, between, uh, what, uh, is happening in the world today and what is, what has happened in the past. There are some differences, but I think the parallels are, uh, indeed striking. Uh, we talk about, um, data revolution, but if I hearken back to the original [00:29:00] information, information revolution, which was.
Guttenberg and the invention of the printing class, the ability to mass produce books of all kinds changed the culture back then because reading was only for the elite. Uh, and this was a really a transformative event around the world and the. The impact it had, uh, on lives of millions was fairly radical.
So Martin Luther, um, was one of the early users of the printing press, and if not for the printing press, uh, the Renaissance would probably have remained an isolated experience in Northern Italy.
Steve Hamm: Yeah.
Neelesh Prabhu: one of the things which we take for granted today, which is the ability for ideas to challenge idea.
That was something that was driven by the, uh, printing press as well, and he called that today the scientific method. So these were really profound and radical changes [00:30:00] introduced by, uh, a new technology. And today we talk about, um, uh, democratization of data or making data accessible to people. So this is not a new trend.
Uh, it has links to the past and it will result in, uh, fairly radical changes, many of which we, uh, can't predict. The other interesting part about that, Steve, is that, uh, you know, while change has remained the constant, the rate at which it's coming to us is, is different. Like it took us 125 years, uh, before, uh, the telephone reached a billion people, and yet it took just eight years for Facebook to do, do just that.
So the, the rate of change. Is not something you've experienced before. But the key takeaway I would say is that we should recognize that we have been here before. This is a, we shouldn't panic, we shouldn't obsess with it, but we should learn from our [00:31:00] history because it can inform us.
Steve Hamm: yeah, yeah, yeah. I, you know, I think it is funny, It's like there're waves within waves. Um, I look back on, I think Bill Gates gave this speech at conduct, I think it was 1990 or something like that. He talked about information at your fingertips. It was a vision for the industry and ever since then, when I.
When I look at, you know, the next 40, 50 years or whatever it's been, well, I guess 30 years, uh, it's really a fulfillment of that vision and we just keep getting more, better access, quicker access, uh, you know, better deeper data, better ability to analyze it, and you see it continuing into the future and, and hopefully, um, it'll make, you know, things better for business and society both and, and individuals as well.
So, This has been absolutely fascinating. I, you know, this whole process of, of getting to know you guys and getting to know your, your, your business is really cool [00:32:00] to me. I mean, I, I like to kind of see who's on the cutting edge of technology and. And DTCC certainly is there, so it's something that I'm gonna be paying it much more attention to in the future since I've just, uh, really learned about the, the organization.
I'm really interested to see what you guys do with blockchain because it just seems, you know, the promise of blockchain people in talking about it for a long time, but it's the kind of applications that you are using it for that really, it seems to me could, you know, be extremely important and, and valuable.
For, for business. So anyway, thank you so much for being on. It's been, it's been a great podcast.
Neelesh Prabhu: Thank you Steve, and it's been great talking to you and I hope, uh, your listeners have found this very interesting as well.