Data Quality Score Calculation

After rules, enforcements and implementations are configured and executed, the data quality score is calculated for each report. This score is based on the failed data as a result of the calculations in the report’s Query fields. This score rolls up across assets to provide insight into data health at multiple layers of a project.

An asset’s data quality score is always calculated based on the average of the total data quality scores for the latest run of all the associated DQ Reports that roll up to that asset.

The data quality score is recalculated when a user:

• Deletes an enforcement

• Deletes an implementation of Method Syniti Cloud DQ

• Changes an enforcement with the Roll Up Data Quality Score checkbox checked

• Changes an implementation with the Include in Data Quality toggle set to on

• Executes a new or existing implementation

The data quality score is not recalculated when a user:

• Changes a rule (incrementing its version)

• Changes an enforcement (incrementing its version)

• Changes an implementation (incrementing its version)

Note

In cases where a data quality score was calculated for an asset but where subsequently there are no implementations rolling up to the data quality score, then the asset’s data quality score is reset to NULL.

Data Quality Score Calculation for Implementations

Implementation data quality score (%) is calculated based upon the number of Errors and Opportunities. The exact calculation is:

Data quality score = (1 - (Error Count / Opportunity Count)) * 100

Example

Implementation 1 has 10 Errors out of 550 Opportunities.

Data quality score = (1 - (22 / 550)) * 100 = 96.36%

Data Quality Score Calculation for Enforcements

Enforcement data quality score (%) is calculated based on the average data quality score of the latest run of implementations assigned to the enforcement. These implementations must have the Include in Data Quality toggle, found on the Create implementation and Edit Implementation pages, set to on.

Example

Enforcement 1 has 2 implementations assigned to it that have the Include in Data Quality Score toggle set to on (Implementation 1 and Implementation 2).

Data quality score = (96.33 + 93.86) / 2 = 95.11%

Data Quality Score Calculation for Rules

Rule data quality score (%) is calculated based on the average data quality score of the latest run of implementations assigned to the rule enforcements. To be included in the score, the enforcements must have the Roll Up Data Quality Score check box checked and the implementations must have the Include in Data Quality toggle set to on.

Example

Rule 1 is assigned to Enforcement 1. Enforcement 1 has the Rollup Data Quality Score check box checked and has 2 implementations assigned to it that have the Include in Data Quality Score toggle set to on (Implementation 1 and Implementation 2).

Data quality score = (96.33 + 93.86) / 2 = 95.11%

Rule 2 is assigned to Enforcement 2. Enforcement 2 has the Rollup Data Quality Score check box unchecked so Rule 2 has no data quality score.

Data Quality Score Calculation for Business Processes

Business process data quality score (%) is calculated based on the rules assigned to it. It is calculated based on the average data quality score of the latest run of implementations assigned to the rule enforcements that have the Roll Up Data Quality Score check box checked and the implementations that have the Include in Data Quality toggle set to on.

Example

Process 1 has 3 rules assigned to it (Rule 1, Rule 2 and Rule 3). These rules have 4 implementations assigned where the enforcement has the Rollup Data Quality Score check box checked and the implementations have the Include in Data Quality toggle set to on (Implementation 1, Implementation 2, Implementation 6 and Implementation 7).

Data quality score = (96.33 + 93.86 + 32.50 + 51.35) / 2 = 68.52%

Data Quality Score Calculation for Subject Areas and Business Concepts (Terms)

Subject Area and Business concept data quality scores are calculated in the same way as business process data quality score.