Anomalies

Anomalies in Qualytics represent data points that deviate from expected patterns or violate defined quality rules, often highlighting issues such as missing values, structural inconsistencies, or incorrect data. These anomalies are detected during scan operations through AI Managed or user-authored checks. When every failed check behind an anomaly is AI Managed, the anomaly displays an AI badge (purple pill with a four-point star icon) — hovering it shows "Identified by AI managed checks."

Anomaly Types

Qualytics classifies anomalies into two types: Record Anomalies and Shape Anomalies. Record anomalies flag rows with data issues like missing or invalid values, while shape anomalies detect structural problems such as missing columns or schema changes. Together, they ensure thorough data quality coverage at both the value and structure levels.

Note

For more information, please refer to the Anomaly Types Documentation.

Anomaly Detection Process

The anomaly detection process in Qualytics ensures data quality by identifying deviations from expected patterns through a structured workflow. It starts with configuring datastores, syncing metadata, and profiling data to understand its structure. Users then apply quality checks—either Authored or AI Managed—during Scan operations. Any failures are flagged as anomalies, enabling timely detection and resolution of data issues to maintain overall data integrity.

Note

For more information, please refer to the Anomaly Detection Process Documentation.