Some of the more common pitfalls affecting the data warehouse are:
Source data is not consistent, is missing or invalid.
Source system data is not usable “as is” in a data warehouse environment, requiring data transformations to be performed.
The source data “meaning” is not fully understood, making the load prone to errors and assessment of data changes difficult.
ETL tool is not utilized to its full capacity.
Source data mappings are not correct.
Business rules implemented for the data load process are not accurate or comprehensive.
Relationships between data from heterogeneous systems are not understood, making the data load prone to errors.
Information stored in the data warehouse is not complete or correct.
End users do not understand the data or the interrelationships between the data.
Without realizing, end users utilize reports for the wrong purpose.
End users validate data warehouse reports against other reports incorrectly.