A data quality dimension is a governance label that describes what kind of data quality problem a validator looks for. Each Validio validator can declare a single dimension and an explicit list of schema fields it covers. Together, the dimension and the field list determine where a validator's results show up across the catalog — in Classifications, on the glossary term data quality tab, and in dimension-based filters on the validators list.

The Six Dimensions

Validio uses a DAMA-aligned set of six dimensions:

Each validator type has a default dimension that pre-fills when you create a new validator. You can change the dimension at any time, and you can leave it unset for validators where dimension is not meaningful (for example, a custom debugging validator).

Data Quality Fields

Alongside the dimension, each validator can declare a list of data quality fields — the schema fields the validator is intended to govern. Data quality fields can differ from the technical fields the validator reads:

• For most validators (Numeric, Null Values, Enum Values, etc.), the data quality fields default to the same field the validator measures.
• For validators that monitor an asset indirectly — for example, a Freshness validator on a loaded_at timestamp that is governing freshness for the actual business field market_value — the data quality fields are the governed fields, not the field the validator computes on.

This separation lets you wire a validator to the right governance surface without distorting its technical configuration.

Set Dimension and Fields on a Validator

The dimension and data quality fields are set in the Details step of the validator configuration wizard, alongside name, owner, and tags.

1. Start the validator configuration wizard on a source.
2. Pick the validator type and complete the metric, threshold, and segmentation steps.
3. On the Details step, the wizard pre-fills:
   • Dimension — based on the selected validator type.
   • Critical fields — (data quality fields) based on the fields you selected for the metric.
4. Edit the dimension or fields if the defaults do not match the governance intent.
5. Click Continue to create the validator.

You can change the dimension and data quality fields on existing validators from the validator update dialog or from the validator details panel.

📘

In batch validator creation (multiple Numeric Statistic or Null Values validators on different fields), the dimension applies to every validator in the batch, but the data quality fields are derived per validator from its own metric configuration.

Browse Validators by Dimension

The validators list surfaces dimension and field tagging in two ways:

• Dimension column — Each validator row shows its dimension as a chip, colored by dimension. Validators without a dimension show "No dimension".
• Filter facets — The filter toolbar includes Dimension and Data quality fields filters, so you can narrow the list to (for example) every Completeness validator on a specific field.

For more on the validators list, see Reviewing Validators.

How Dimensions Roll Up Across the Catalog

Dimension and field tagging is what makes coverage visible elsewhere in the platform:

• Classifications — A Critical Data Element declares its required dimensions. A field tagged with the CDE is considered covered for a dimension when a validator on that field declares the matching dimension. Required dimensions without a covering validator surface as gaps.
• Glossary term details — The data quality tab on a glossary term aggregates the dimensions of validators on every field assigned to the term. The colored dots on the glossary list and term details give an at-a-glance view of which dimensions are tracked, missing, or not applicable.
• Schema fields list — The Schema & Profiling tab supports filtering by data quality dimension, so you can find every field that lacks coverage for a given dimension.
• Validator recommendations — The recommendation agent uses dimension and field intent to propose validators that fill governance gaps (e.g., "completeness checks on customer_id" produces a Null Values validator pre-tagged with the Completeness dimension and customer_id as the data quality field).

Tune the Passing / Warning / Failing Thresholds

Dimensions roll up into a per-dimension health state — Passing, Warning, or Failing — based on the underlying validator quality scores. The thresholds for those bands are workspace-level and are configured on the Workspace > Settings page. For more information, see Configuring Global Settings.

Related Resources

• Configuring a Validator
• Managing Validators
• Reviewing Validators
• Classifications
• Business Glossary