AFTAs 2017: Best Data Management Initiative—Deutsche Bank

DQ stands for “Data Quality” and that’s what this initiative is all about—a project aimed at creating a common language to describe data problems in order to improve the firm’s data on a global basis. “It creates a common platform as a catalyst to allow people to collaborate and talk in a common language—that’s the number one thing,” says James White, director of Deutsche Bank’s Data Quality Management unit, which is overseeing this project. “It takes disparate problems and issues and creates something that’s joined,” he says.

While 2016 was more about proving the worth of the project and setting the strategy, 2017 was about rolling out DQ Direct globally. This initiative—and the underlying platform—aims to improve data quality by transforming the way that the bank’s 100,000 employees think, feel and act toward its data. As such, there are six steps involved in this change of culture. First, there’s initial training as to the “whats and whys”. Then they identify a DQ Direct “champion”. From there, they gain feedback regarding the quality of issues loaded into the tool. Next, there’s a presentation of management information. After that, they confirm that all known and well-formed data quality issues are loaded into DQ Direct. And, finally, there’s a presentation of data quality issues to the data governance council.

Over the last 12 months, Deutsche Bank has completed the six steps across 56 of its 59 coverage areas—which is not a legal designation, but rather a way that the bank views its various operations, from business areas, products, processes to regions and key projects. 

Since its launch two-and-a-half years ago, 7,800 data issues have been registered; 4,900 data issues have been consolidated; 1,100 data issues have been resolved; and 1,800 issues remain open and visible, each with a descriptive business case allowing prioritization for remediation. 

For 2018, the bank will finish rolling out DQ Direct to the other three outstanding coverage areas. Beyond that, the Data Quality Management team will look to develop a way to better measure performance versus expectations so they can better show the impact of the project, directly, and so that people can “tell a story” about the data issues they encounter.