Data management technology provider of the year: State Street

Repurposed internal tool builds on custody data already on file

subbiah-subramanian
Subbiah Subramanian, State Street. Image: www.studioeleven.com

Buy-side Awards 2016

Buy-side firms want their business experts to spend time extracting value from data, not gathering and preparing it. So why not hand that over to an organisation that has the infrastructure and experience to do it more efficiently?

Investment managers already trust custodians with their assets, so why not their data, especially where the custodian holds a large part of it already?

With these assumptions as its starting point, State Street, based in Boston, Massachusetts, took data management capabilities it developed for internal purposes and packaged them as an external standalone service for the buy side.

The firm began offering data management services via its enterprise service platform to existing middle-office clients in 2009, with 13 firms using the service by 2014.

"Then we had the idea of offering it beyond our middle-office clients. By then, the product and its operation was mature, so we decided to lift it out and offer it to all our clients, regardless of how much or little business they might have with State Street," says Subbiah Subramanian, senior vice-president at the firm.

Building clients

This data-as-a-service offering – DataGX – now has three clients live, with a handful of others onboarding and more in the pipeline. The service brings together three main services required by buy-side firms.

First, there is the client-specific custodial data, such as holdings, end-of-day positions and corporate actions. "This data typically originates from the client's middle office and managing it includes aligning the data and validating it. We make sure that if a client holds, for example, Microsoft stock, it appears in one place, even though the data might come from five different sources. And we make sure that when we get data from a custodian, it doesn't suddenly jump in value, or if it does, there is a good reason," says Subramanian, who is head of DataGX North America.

Second, there is the loading and management of market data, which can include prices, indexes and yield curves. "Clients often want to augment their data with other sources. We make sure that, for example, reference data from a custodian matches that from a data vendor, so the client has one consistent set of information," he explains.

The third element is data the client itself produces, such as internal credit ratings or other risk data. "Clients tell us they don't want to create data islands in their organisations, so they want to load this data onto the platform as well," says Subramanian.

We make sure that if a client holds, for example, Microsoft stock, it appears in one place, even though the data might come from five different sources. And we make sure that when we get data from a custodian, it doesn't suddenly jump in value, or if it does, there is a good reason
Subbiah Subramanian, State Street

Once all the data is on the platform, it can be aggregated and redistributed for clients' investment analytics and reporting, including regulatory reporting. Aggregation can be by any factor the client chooses, such as country or sector of exposure or duration buckets. Best-practice data governance principles are applied in terms of data quality, lineage tracking, security and controls.

"If we didn't provide this service, then clients would have to have the software and operations to do it for themselves. We say: ‘We'll take on the data management, and you can focus your people and resources on using the information effectively,'" says Subramanian.

DataGX remains part of State Street's enterprise service platform, and benefits from the continuing development of the infrastructure for other users. "State Street uses its platform as a way of getting data out of our legacy systems into a standardised business format. We have over $27 trillion of assets in custody, so we see a lot of enhancements coming out of the internal requirements. That all goes into the same platform because it is a single code base, so DataGX clients get the benefit as well," he says. The data itself sits on a high-performance database machine, and is replicated across three data centres for reliability.

Extending the service

State Street is now looking at how it can extend the service, including the possibility of bringing unstructured data into the platform. "Companies come to us and say they can scrape data from various social media sources and financial information services. We are looking into how we might incorporate this public unstructured data so our clients can benefit from it," says Subramanian.

State Street is also researching how it might use machine learning to enhance its data management. The firm is already applying machine learning to other parts of its services: "We are looking to see how we could use the technology to improve the management of our structured data, so we don't have people scrubbing the same data over and over, but bring some learning and automation to the process."

State Street also offers risk and performance analytics and compliance tools as part of its Global Exchange suite of services, and is working on the ability to plug these tools into DataGX.

"We already have clients' data, so we could provide risk, performance and compliance services with the assurance that the same set of information is being used across all these areas," says Subramanian.

Only users who have a paid subscription or are part of a corporate subscription are able to print or copy content.

To access these options, along with all other subscription benefits, please contact info@risk.net or view our subscription options here: http://subscriptions.risk.net/subscribe

You are currently unable to copy this content. Please contact info@risk.net to find out more.

You need to sign in to use this feature. If you don’t have a Risk.net account, please register for a trial.

Sign in
You are currently on corporate access.

To use this feature you will need an individual account. If you have one already please sign in.

Sign in.

Alternatively you can request an individual account here