Enterprise Data Warehouse Cost of Use
BACKGROUND:
The U.S. Postal Service created the Enterprise Data Warehouse (EDW) as the main source for storing data across functional areas throughout the organization. The EDW stores data from about 100 systems or applications. The process for loading data from these applications into the EDW is referred to as an extract, transform, and load process.
There are about 14,000 individual programs needed to perform this process. The extract, transform, and load process requires a series of tasks that are accomplished with contractor support. Additional contractor support has been required because of the growth in the number and complexity of manually developed extract, transform, and load scripts. Replacing the current method of coding and maintaining the extract, transform, and load scripts with a tool that automates the initial build of the scripts would reduce both current and future maintenance and development costs.
Our objective was to evaluate whether the Postal Service could reduce EDW costs by implementing technological advances available in the marketplace
WHAT THE OIG FOUND:
Opportunities exist for the Postal Service to reduce EDW costs by implementing technological advances available in the marketplace. Commercial software is available to automate the interface process and replace the current manual process. As a result, the Postal Service spent $500,000 in labor costs that it could have avoided in fiscal years 2012 and 2013. In addition, the Postal Service could save about $1.9 million annually in future years. Further, standardizing the process would eliminate the potential for manual programming errors and allow administrators easier access to data from the various functional areas
WHAT THE OIG RECOMMENDED:
We recommended the Postal Service acquire and implement a tool that automates the scripts used to perform the extract, transform, and load process.