With data at the core of the industry’s efforts to recover from the impact of the Credit Crunch, it’s imperative that the platforms used to manage a financial institution’s information services are able to draw upon established business processes in order to reach the Holy Grail of business process optimization. That, according to Nigel Matthews, global head of product management and strategy, for Thomson Reuters Enterprise Platform for Data Management, was a key consideration as the company set its strategy for its enterprise data management strategy.

The company took at its starting point two sets of market research. The first, conducted by A-Team Group for Reuters in 2007, suggested that data consumers found that obstacles including data complexity and cost made it impossible to optimize. A further review of the research with clients found that many believed that the extension of data platforms into the middle and back office, a consistent approach to dealing with data convergence and the automation of key functions like security set-up and identifier creation, could help.

The Thomson Reuters team also referred to research by the Enterprise Data Management Council that tracked the five stages of data management maturity: from initiation, to realization, to definition, management and optimization.

With this in mind, Thomson Reuters defined its enterprise data management technology stack in seven layers. Starting with data storage as the commoditized bottom rung, the layers run up through data modeling and normalization, data acquisition and validation, data cleaning and rules management, data operations workflow, metrics and reporting and finally data quality optimization. According to Matthews, the lower tiers of this stack present opportunities for commoditization, while the middle tiers often represent obstacles in reaching the goal of data quality optimization.

This EDM approach doesn’t offer its full value, however, unless it’s combined with steps toward business process optimization, he said. Here, he outlined a seven-tier middleware stack, starting with the commoditized bottom tier of messaging, and ranging up through message brokering, application integration, business process management, business operations workflow, business activity monitoring and business process optimization.

To pull the two stacks together, he said, Thomson Reuters added a so-called integrated semantics governance component, which binds appropriate elements of each stack in order to meet the objective. This, he said, could result in the creation, for example, of a canonical data model for all data exchange across the platform, allowing data convergence to meet business challenges.