Skip to main content
Validations are rules applied to your datasets that measure specific metrics and flag anomalies. For example:
  • Count total transactions in the payments table every day
  • Track the percentage of duplicate entries in the customer_email column
Validations can be created manually, suggested automatically when a dataset is discovered, or generated by the Data Validation agent.

​
The validations list

Navigate to Validations in the left sidebar to see all validations in your workspace.
ColumnDescription
NameValidation identifier.
TypeThe validation category (e.g. Completeness, Uniqueness).
StatusCurrent state — Healthy, Alerting, or Failed.
Last ResultsVisual indicator of recent run outcomes.
Last EvaluatedTimestamp of the most recent execution.

​
Statuses

StatusMeaning
HealthyThe metric is within the defined threshold.
AlertingThe metric has crossed the threshold and triggered an alert.
FailedThe validation run encountered an error.

​
Filters

Use the filter bar to narrow the list by:
  • Status — Healthy, Alerting, or Failed
  • Category — Validation category group
  • Method type — The specific check method
  • Threshold type — Static, dynamic, or anomaly-based
  • Collections — Custom groupings of related validations
  • Data source — The connected database
  • Search — Filter by name or run ID

​
Actions

Each validation row supports the following actions:
  • Run — Execute the validation immediately
  • Pause — Suspend scheduled execution
  • Edit — Modify the validation configuration
  • Delete — Remove the validation
  • View details — Open the full results and metrics view

​
Collections

Group related validations into collections to run them in parallel or organize them by domain. Select one or more validations and click Add to Collection.

​
Create a validation

Click New Validation to configure a new check. See Configure Validations for a step-by-step guide.