8. Default rules - making do with bad data. Existing source system data will doubtless violate some quality rules. It is not feasible to wait for the source system to fix these problems, as they will likely be undergoing other changes simultaneously. Institute a process for setting up default rules that will be used if these quality violations are observed in data. As this is also a business requirements process, it is critical to set expectations for these kinds of rules with the business requirements teams.
9. Use the right technology. Warehousing technologies are becoming ever more sophisticated and scalable, but many technologies still place severe limits on the amount of data that can be held in the warehouse and on query volume or other metrics. Pick a technology that can scale properly to the expected volume of data and broad usage. Warehouses have a way of becoming too successful, leading to scalability issues and ultimate user disillusionment.
10. BI tools are distraction. Many teams spend far too much time focusing on business intelligence tools. The RDW provides massive benefit simply by aggregating data, so flashy BI tools are not really necessary to score wins with users. The user population is likely to be small at inception (a half-dozen users for the whole warehouse is not uncommon), so scalability of data is not as important, either.
A risk data warehouse also needs to be flexible and available because users need to discover uses for data by studying this in an unstructured fashion. A relatively amorphous environment is therefore best at the inception. Once there is a buy-in from business users, there will be plenty of time (and money) to apply sophisticated BI technology on top of the warehouse.
Building a risk warehouse is an intimidating task, especially in a time-challenged regulatory environment, but it's not an insurmountable one. By using proper project management techniques and employing adequate resources, in conjunction with the methods described in this article, it is eminently feasible to build a robust RDW that can generate tremendous goodwill and become a significant corporate asset with business-transformational potential.
References:1. These are the so-called G-10 countries: Belgium, Canada, France, Germany, Italy, Japan, the Netherlands, Sweden, Switzerland, the United Kingdom, the United States and Luxembourg.
Dilip Krishna, CFA, consults on risk management in financial institutions and heads NCR Teradata's enterprise risk management practice. He has previously been involved with several large implementations of financial and trading systems in large North American institutions. He can be reached at dilip.krishna@ncr.com.
