DQC Logo
|

Segmentation for data quality checks

Segmentation allows you to apply a data quality rule per group, rather than across the entire table. It works similarly to the GROUP BY clause in SQL and is especially useful when you want to validate conditions within each segment, such as per region, business unit, or product category.

What does segmentation do?

With segmentation, the DQC Platform applies the rule individually to each group defined by the selected column(s). This means:

  • The condition is checked per segment

  • You can detect localized issues (e.g. only one department has outliers)

  • You retain granular visibility instead of losing detail in aggregated checks

Outlier rule applied per “Department ” rather than across all departments

This is how to set up a rule with segmentation:

When to use segmentation

Use segmentation when:

  • You want to apply the same rule logic independently per group

  • You suspect variation across business units, teams, or categories

  • You want to reduce false positives caused by aggregated outliers

Example use case

Scenario: You want to apply a “No outlier” rule to a numeric field, but values vary widely across departments.

Solution:
Add a segmentation by Department Code:

  • Rule: No outlier on Awarded Amount

  • Segmentation: Department Code

The rule now checks for outliers within each department, not across all rows

Compatible rule types

Segmentation is currently supported for:

  • No outlier

  • Text pattern

  • Categorical

Configuration

  1. Create or open a supported rule

  2. In the Segmentation section, select one column

  3. Save the rule — it will now apply per group rather than globally

Tip: Combine with filters

Segmentation and filters can be used together to scope rules precisely.
Example:
Filter: Remaining amount > 5000
Segmentation: Department
Rule: Awarded Amount has no outliers

undefined Notes

Segmentation for data quality checks | DQC