ETL (Extraction, Transformation and Loading of Spend Data)
A federal agency may have multiple data sources from where it needs to aggregate data. A detailed spend data audit is required to list the number of systems from where relevant spend information can be extracted.
Some examples of data sources in federal agencies are
Federal Procurement Data Systems
Procurement Credit Cards
Suppliers
Inter-agency contract systems
General Ledgers
Acquisition Plans
The challenges in data aggregation can range from a simple extract from one procurement system to very complex mapping for organizations (such as DHS) where there has been a recent merger of several agencies under one large umbrella.
Data extraction and aggregation is carried out using standard industry tools available from companies such as Informatica or Ascential (now IBM). These tools plug into industry standard ERP systems and e-Procurement systems and pull data from different source files, mapping them into single common file format prior to loading on a spend cube. Normalization and mapping functions are carried out within these tools. In case of non-standard (proprietary systems), data extraction routines are written by Zycus partners (such as IBM), to facilitate data extraction and normalization prior to loading the data on a zycus spend cube.
ETL (Extraction, Transformation and Loading of Spend Data)
