Numeric Checks

Average, Minimum, Maximum, Sum tests.

These four checks compute a single aggregate statistic — the average, minimum, maximum, or sum — of a numeric column, and check that it falls within an expected range. They only apply to numeric column types (integer, float, decimal, and similar).

When to use it

• A business expects the average order value in an orders table to stay between $20 and $200. A daily average check flags a sudden drop, which could indicate a currency or unit-conversion bug introduced upstream.

• A business rule caps a single order at 100 units. A daily maximum check on the quantity column catches data-entry errors or abnormal orders that exceed the cap.

What's configurable

• Schedule — a fixed recurring schedule, or Manual (triggered on demand). If Manual, the monitor must use Full Table Scan.

• Scan Method — how much of the table is read each run:

  • Full Table Scan — reads the whole table every run.

  • Incremental Scan — reads only the rows added since the last run, based on a timestamp column or a custom filter expression.

• Threshold — a Numeric Range: a minimum and/or maximum for the computed statistic. At least one bound is required; bounds are inclusive.

Worked example — Average

• Asset: orders table, column order_value

• Monitor type: Average

• Scan Method: Full Table Scan

• Schedule: Daily

• Threshold: Numeric Range — min 20, max 200

Resulting rule: "An incident will be created when the average of order_value is below 20 or above 200."

Worked example — Maximum

• Asset: products table, column quantity

• Monitor type: Maximum

• Scan Method: Full Table Scan

• Schedule: Daily

• Threshold: Numeric Range — max 100 only (no min)

Resulting rule: "An incident will be created when the maximum of quantity is above 100."

Notes

• Changing the schedule or scan method later is a destructive change (it creates a new monitor version). Adjusting only the threshold value is not.