Accuracy Checks

Accuracy checks validate statistical properties of numeric columns to ensure values fall within expected ranges and meet business expectations.

Follow these best practices for accuracy checks:

  • Set meaningful thresholds based on historical patterns in your data.

  • Configure both Minimum and Maximum checks where applicable to catch anomalies in both directions.

  • Use standard deviation checks to identify statistical outliers.

  • Consider seasonality when setting thresholds for business metrics.

  • Update thresholds periodically as business conditions evolve.

Available Metrics

  • Minimum / Maximum: Validates value boundaries within acceptable ranges.

  • Average: Ensures mean values meet business expectations.

  • Sum: Validates aggregate totals for financial or quantity data.

  • Standard Deviation: Monitors data distribution consistency.

  • Variance: Tracks statistical variance within acceptable bounds.

Configuration Examples

Check Description

Configuration Steps

Example Check

Ensure product prices are positive

  1. Select price column.

  2. Choose Min check type.

  3. Select >= operator.

  4. Set threshold to 0.

min(price) >= 0

Validate reasonable customer ages

  1. Select age column.

  2. Choose Max check type.

  3. Select <= operator.

  4. Set threshold to 120.

max(age) <= 120

Ensure average order value is within normal range

  1. Select order_total column.

  2. Choose Average check type.

  3. Select between operator.

  4. Set threshold to 50 and 500.

avg(order_total) between 50 and 500