Cube_Data_Quality holds details of records with data quality issues. It can be used both for reporting on current records and for showing trends over time.
Dimensions in the cube based upon Shared Dimensions are as below. Details of these columns are in the Shared Dimensions section.
|Shared Dimension Name||Name in Cube||Notes|
DQ Check Detail
This contains various columns from the DQ_Check_Detail table, which holds the records failing the data quality checks.
|Check Field Name||Name of field(s) on which the check was carried out|
|Check Field Value||The value(s) of the fields which failed the check|
|DQ Category||Higher level category, allowing grouping of checks by team/ process as required.|
|DQ Check Name||Unique name for the check|
|DQ Impact||Indicator of how serious the failure is.|
|DQ Note||Extra information about the record, to help prioritise/ find as required.|
|Key Field Name||Name of field(s) which can be used to identify the particular record.|
|Key Field Value||Value of the above field(s)|
|Source System||System where the records failing the check some from.|
|View Name||Data Quality view which contains the check against.|
|DQ Run Date||Date the check was run|
|DQ Current Record||'YES' for current record, based upon latest batch. 'NO' for record created in previous batch.|
|Cube Measures Group||Cube Field||Derivation||Description|
|DQ Check Detail||Count DQ Recs||Number of records.|
The table DQ_Check_Detail is populated based upon views in DW beginning 'DQ_'. These are all the same structure, enabling a script to iteratively load records from each table as part of each refresh, and for new checks to be added simply by creating a view meeting the above criteria.