Columns
Industry Insights
Avoid the pitfalls
Logical data models are key to successful risk management for financial institutions.
by Sam Harris
Effectively and transparently measuring and managing risk have always been critical precepts of financial institutions. But never before has access to timely and accurate risk-related data played such a life-or-death role in those organizations’ operational viability and long-term stability.
These are unprecedented times for financial institutions that have faced record losses, billions of dollars in forced write-downs and failure at unforeseen rates.
Financial leaders are quickly realizing that risk information is their most valuable asset for navigating these uncertain times. The institutions that have the clearest picture of the risks they face and the choices before them are most likely to succeed. It’s also extremely important to promote high transparency to help improve regulatory compliance and boost customer and stakeholder confidence.
Implementing a detailed, enterprise-wide logical data model (LDM) is crucial to ensure that financial leaders have all of the information they need to understand what risks exist. With this knowledge, they can make informed decisions about how to manage those risks and keep their institutions thriving.
An LDM for better insight
Like the blueprint for a building, an LDM defines the structure of all of the data for an enterprise. Having only one logical model will ensure consistency and provide a single source of truth within the enterprise as it charts the relationships of data resulting from organizational events. By fostering consistency in language and concepts across the organization, an LDM enables better capture of all pertinent risk information. Analysis of that data then leads to quicker and smarter decisions.
The LDM is often a first step toward creating an enterprise data warehouse (EDW) that can quickly and easily access and analyze risk data from disparate sources across the organization. With this information at hand, users can better determine risk exposures and make crucial decisions in a timely manner.
For example, officials at a bank learn at 3 p.m. that a commercial customer is on the verge of insolvency. The bank must determine its exposure to that company as quickly as possible, so it searches data stored in numerous systems for exchange traded exposures and spreadsheets for over-the-counter instrument exposures.
Not only must officials discover the amount of direct exposure they have from the potentially soon-to-be-insolvent company, but they must also determine the status of any relevant guarantees, collateral or related credit enhancements. Without an enterprise LDM and EDW to collect and organize all of the data, the officials would have to manually piece together what information they could glean from disparate systems—wasting valuable time and yielding sub-optimal results.
A logical model facilitates financial risk management as it:
- Improves the categorization of risk information and risk-measurement and -management capabilities
- Facilitates benchmarking the current state of an organization’s risk information against best practices
- Allows organized growth of the data warehouse for more than risk management uses, over time, without having to re-architect
- Provides discipline and structure to the complexities inherent in risk management data
- Facilitates communication with consistent terminology among business units to ensure company leaders get the risk information they need in time to act
Beyond the LDM
Creating the logical model begins the journey toward better risk management. The next step is to build a physical data model that precisely implements the structures and foundations outlined in the LDM.
The quality of the physical data model is a direct reflection of the quality of the LDM, so it’s important that the logical model be as accurate and detailed as possible. It must depict not only data inputs and outputs but also the relationships among the data. Furthermore, a good LDM is nearly “future-proof”—that is, the model is flexible, extensible and scalable, with the capability to grow and change with an organization, no matter what surprises or uncertainties the future holds.
By fostering consistency in language and concepts across the organization, an LDM enables better access to risk information.
Because of the inherent complexity of a logical model, financial leaders should consider finding a partner with best-practice experience creating them for financial organizations. Companies that try to create their own models from scratch invariably increase their overall project risk and implementation time. Additionally, they miss the opportunity to benefit from the lessons learned by organizations that have built LDMs before them.
When choosing a partner, look for a company that offers risk-reducing project accelerators such as a pre-built logical model for financial services or insurance firms that can serve as a baseline. Make sure the company has the expertise required to modify that baseline model to reflect the unique needs of your organization.
Potential benefits
Financial leaders can realize numerous short- and long-term benefits from creating a logical model:
- The ability to access information across the enterprise increases with shared terminology.
- Consistent reporting and analytics boost transparency internally as well as externally. This can help organizations make smarter strategic and tactical decisions, in addition to quickly demonstrating audit and regulatory requirements more easily.
- Better data consistency and accuracy improve operational efficiency while reducing risk.
- Eliminating the technical and human redundancy inherent in siloed systems as well as in data marts reduces expenses.
Creating a detailed logical model is an indispensable step for financial institutions wanting to maintain solid footing in shaky financial times and beyond.
Sam Harris, director of Enterprise Risk Management, leads Teradata’s global ERM program for the banking, capital markets and insurance industries.
Photography by Corbis
