Skip to content

Data Validation – Rule-Based Checks

AI-Driven Data Validation Module for Data Quality and Observability – digna

Purpose

The Data Validation module ensures the quality of data through precise, rule-based checks.
It allows organizations to define deterministic business and technical validation logic, ensuring that data meets compliance standards, contractual SLAs, and regulatory requirements.

By combining in-database rule execution, complete audit trails, and integration with other digna modules, Data Validation guarantees consistent and traceable Data Quality and Observability across complex enterprise environments.

Technical Overview

Supported Validation Types

  • Equality Checks
    Confirm that values match expected results (e.g., reference codes, Boolean flags, categorical mappings).

  • Thresholds & Ranges
    Validate numeric measures or KPIs against defined limits — static or dynamically derived.

  • Reference Lists & Lookups
    Check whether field values exist within approved master data sets (e.g., VAT codes, ISO country lists, product catalogs).

  • Cross-Column Consistency
    Ensure relational correctness (e.g., currency aligns with region, risk category matches asset type).

  • Null Handling Rules
    Detect unexpected null or empty values in critical columns.

Execution and Logging

  • In-Database Processing – All validation rules execute directly in your database (Teradata, Snowflake, Databricks, PostgreSQL, etc.).
  • No Data Extraction – digna never transfers raw data outside your environment.
  • Full Traceability – Each rule result is logged with timestamp, responsible dataset, record counts, and pass/fail outcomes.
  • Audit