It's no secret that capital markets and investment management firms need better ways to aggregate, analyze, and normalize data, both structured and unstructured. "It's not an overstatement to say things are still mostly a mess," says Gert Raeves, research director for the CEB TowerGroup.
Whether it's meeting compliance requirements, gaining competitive advantage, or doing a combination of the two, Wall Street firms are seeking enterprise data aggregation remedies. Many are exploring ways to enable every department and discipline -- from fixed income and equities to risk management and compliance -- to tap into and utilize the same data sets in real time.
While it's tempting to assume that technology can supply all the answers, experts say otherwise. "Historically, the front-office product focus has created silos that extend throughout the product chain," Raeves points out. "Further, as firms grow geographically, or by acquisition, this creates operational silos. From these silos, data silos naturally result."
Consequently, the challenge of enterprise data aggregation goes beyond implementing technology. It's equally about formulating -- and implementing -- a strategy to address these historical organizational and cultural silos. "Although technology is a component, the problem is also business and cultural," asserts Virginie O'Shea, senior analyst at Aite Group.
And it's precisely this step where many Wall Street firms have been mired in inertia. According to CEB TowerGroup's latest data management maturity survey, nearly half of respondents either didn't have a data management plan or hadn't started executing on an existing plan.
Research by IDC Financial Insights concurs. "In our 2011 sampling of chief risk officers, seven in 10 stated their risk information 'underperformed' against expectations," says Michael Versace, global research director for the firm. "Fast-forward to December 2013 and Basel Committee on Banking Supervision's progress report showed things haven't changed much."
Stop playing hide-and-seek
For many capital markets firms, just knowing where to find data -- much less aggregating it -- is a fundamental issue. "Business functions have different taxonomies and different tagging strategies," says O'Shea, who is working on an industry data aggregation report due out later this year. "It's especially problematic where systems are decades old."
This issue is leading many firms to tackle governance as the logical place to begin their data aggregation initiatives. "Firms are actively investing in ways to measure data quality and establish methods for improving it at the enterprise level. After all, you need to understand your data before implementing technology to automate the aggregation of it."
Accompanying this trend is a change in corporate accountability for data governance. "We're seeing a shift in responsibility away from the CTO and IT," she says. "Instead, accountability is moving to the COO and business functional heads."
"It's an effort to get the people who input the data to adhere to the definition, taxonomies, and tagging. The goal is eliminating all the cleanup at the back end."
Strong leadership = success
Regardless of who is charged with data governance, experts stress successful enterprise aggregation efforts begin with committed leadership. "It's imperative to have the right tone at the top," says Versace.
Leadership must also embrace data aggregation as an ongoing journey, says Raeves. And it must commit to realigning the corporate culture to match. "By the time you finish aggregating one area, a new product or specialization will have created a new flavor of a fragmentation," he says. "Or a new regulation will create new reporting needs. So it's important to develop an aggregation mind-set as well as a strategy that you execute again and again."
Functionally, Versace favors using a shared services model as a mechanism to institutionalize aggregation and share costs simultaneously. "For example, if you establish a golden source of truth for certain liquidity data that's required for quarterly reporting," he suggests, "then the efforts to create that source can be charged back just like any other IT or operational function."
Adopt modern tools
From a technology deployment perspective, experts are seeing two common approaches to modernizing data and information management. "Firms are either adopting enterprise data management systems or distributed data platforms," O'Shea says. "The former creates a single data pool, while the latter integrates pools of data that exist across the firm."
What's more, notes Versace, open source technologies such as Hadoop are not only attractive, but are also now being commercialized by all of the leading technology vendors. "These technologies assist with quickly mining, gathering, and serving up large amounts of data for analysis."
In addition, interest in a new breed of risk and compliance pure-plays is growing. "These solutions can actually identify relationships between data," he says. "Such applications can track the life cycle of key data elements from source systems to all of the reporting and use channels of the organization."
It's also critical to pursue next-generation data and information visualization technologies. "They not only aggregate data and present a visual overview, but also provide drill-down capabilities," Raeves says. "We see these technologies as the key opportunity space for firms to effectively manage the volume and complexity of data sets."
Regardless, experts stress looking beyond compliance as a motivation for aggregating data. "After all, if a firm does it right, it can actually get better at measuring its potential to sell to a customer across multiple business lines," Raeves says. "So, for smart institutions, the business benefits are as important as the initial compliance driver."Anne Rawland Gabriel is a technology writer and marketing communications consultant based in the Minneapolis/St. Paul metro area. Among other projects, she's a regular contributor to UBM Tech's Bank Systems & Technology, Insurance & Technology and Wall Street & Technology ... View Full Bio