You are here: Processor Library > Profiling > Record Completeness Profiler

Record Completeness Profiler

The Record Completeness Profiler allows you to get an overview of how complete, or otherwise, your data is. It informs you how many of the attributes that constitute the record contain data, and bands the records according to the number of complete attributes.

Use

Use the Record Completeness Profiler to find dummy records that have been entered without capturing the proper information. When this occurs, users usually enter the minimum amount of data that they have to, meaning the record will be less complete than a full valid record.

Configuration

Inputs

Any number of attributes required in completeness analysis.

Normally, you would analyze completeness across all the original attributes in your data.

Options

None

Outputs

Data attributes

None

Flags

Flag attribute	Purpose	Possible Values
PercentPopulated	Indicates the percentage of the selected attributes which are populated	Number between 0 and 100
PopulatedAttributes	Indicates the number of the selected attributes which are populated, out of the total number of attributes selected.	'Number' of 'Number'

Execution

Execution Mode	Supported
Batch	Yes
Real time Monitoring	Yes
Real time Response	Yes

Results Browsing

The Record Completeness Profiler produces a summary view of its results, showing the following statistics:

Statistic	Meaning
Record completeness %	Each distinct percentage completeness found, calculated in terms of the percentage of attributes analyzed that were not null.
Complete attributes	How many attributes were not null, and how many attributes were analyzed.
Matching records	The number of records with the given percentage completeness.

Note:The Record Completeness Profiler assesses whether each attribute value is Null or Not Null. Empty Strings, or values that contain only spaces or other non-printing characters are, by default, converted to Null values in the Reader, so that you have a consistent view of whether or not each value contains any meaningful data. However, if such No Data values are not converted to Nulls in either the snapshot or the Reader, they will be considered as 'complete'.

Example

In this example, the Record Completeness Profiler assesses completeness across 4 attributes in a table of Customer records, producing the following summary results:

Summary view

Drill down to find records with a specific level of completeness. For example, to see the record above with 75% record completeness drill down on that row of the summary grid.

Drill Down View