Coverage health

To understand data health, you need to know more than just whether tests are passing or failing. You also need to know what you're testing and how often.

Coverage health metrics

To help you have a better understanding of data health, GX Cloud provides the following coverage health metrics on the Data Assets page:

• Active Coverage: The percentage of Data Assets that have been validated in the last 30 days with an Expectation for schema, volume, or completeness. This is calculated as:

  Formula

  ((% of Data Assets validated for schema) +
   (% of Data Assets validated for volume) +
   (% of Data Assets validated for completeness))
  / 3

• Active Data Assets: The percentage of Data Assets that have had any Validations in the last 30 days. This metric does not consider what kinds of Expectations have been validated.

• Coverage for the following data quality issues. Note that these metrics consider only whether or not Expectations exist. The following metrics do not consider whether the Expectations have been validated.

  • Schema: The percentage of Data Assets that have at least one schema-focused Expectation. This includes the following Expectations:
    • ExpectColumnToExist
    • ExpectColumnValuesToBeInTypeList
    • ExpectColumnValuesToBeOfType
    • ExpectTableColumnCountToBeBetween
    • ExpectTableColumnCountToEqual
    • ExpectTableColumnsToMatchOrderedList
    • ExpectTableColumnsToMatchSet
  • Volume: The percentage of Data Assets that have at least one volume-focused Expectation. This includes the following Expectations:
    • ExpectTableRowCountToBeBetween
    • ExpectTableRowCountToEqual
    • ExpectTableRowCountToEqualOtherTable
  • Completeness: The percentage of Data Assets that have at least one completeness-focused Expectation. This includes the following Expectations:
    • ExpectColumnValuesToBeNull
    • ExpectColumnValuesToNotBeNull

Only current Data Assets are considered in coverage health metrics. Deleted Data Assets are excluded from the calculations even if they've had Validations within the last 30 days.

Next steps

• If Active Data Assets are low, schedule recurring Validations.
• If Schema, Volume, or Completeness coverage is low, add Expectations.
• When adding new Data Assets, automate standard data quality rules.