Michael Shashoua: Data-Handling Storms

With a swirling, confusing array of data management solutions being offered to address the woes of compliance with identifier standards, the data needs for capital adequacy and risk management rules such as Basel III and Solvency II, and plain old efficient and accurate data handling, participants at last month’s Toronto Financial Information Summit wondered how to make data sorting at their firms meaningful, and if data management techniques might supersede data handling methods to that end.

Answers?
Are analytics the answer? And if so, for which asset classes? These questions were raised by professionals from Bank of America Merrill Lynch (BAML), the Ontario Teachers’ Pension Plan (OTPP) and the Information Systems Audit and Control Association (Isaca). The prevention of several risks—operational risks, fraud, inappropriate access, and money laundering—can depend on the same subsets of data, says Alexander Ambramov, director and corporate relations committee chair at Isaca. There can be non-traditional audiences for certain data sets, and non-traditional tools for managing data that some audiences may use, he added.

Data visualization could provide an answer, but should begin with something like a “Google for the enterprise,” says Mark Blair, director, fixed income and alternative investments at the OTPP. “Before we visualize the data, we have to actually get the tool in there that makes the network connections.”

An equities group may have information about a company’s equity, while information about the same company’s debt resides in a fixed-income group, and there is no easy way to link those, Blair says. Linking that information could possibly be done by connecting the people and relationships your firm may have with such a company. “Someone may have approached us through different channels,” he says. “Did we find out something that would be relevant to today’s decisions?”

The use of analytics can bleed into real time and working with market data, but managing the historical data is the basis for trading decisions. Gaining insight into sophisticated black-box models and how they manage data for real-time analytics and trading is a rarity unless you’re a potential investor, says Jacqueline Allen, director of execution services at BAML. Finding the meaningful data hinges on having comprehensive connections of all the relevant data, on the securities themselves and the parties involved in trading them, or investing in the companies. Having adequate sourcing is the prerequisite for being able to sort data effectively.

Management, Not Process

Could the means of managing and organizing data, rather than the ways used to view it, be the solution to get meaningful use out of data? In the past year, customers of enterprise data software provider GoldenSource have been asking more often about best practices and data governance, according to Steve Engdahl, senior vice president of product strategy at the company. “It’s very clear that there’s a lot of green field out there, and everyone is on a journey of establishing data governance at the cross-silo level,” he says.

The regulatory pressures on the industry, as well as the velocity of changes happening that concern the use of data in the industry, have shifted the models for the use of data and changed the “journey of data governance,” says Adele Pugliese, director of data governance and management at Scotiabank. “We have to realign the model with the business strategy. Data governance has a very strategic role within our organization and sets the foundation for policy to help the organization achieve its goals. It’s responding to the changes that are occurring.”

While much has been addressed and considered when it comes to identifier and capital adequacy standards, and the need for data to support compliance, there appears to be quite a lot of thought under way on how to practically manage data and work with it more efficiently.