Attribute - DQLabs PRIZM

Below is a detailed reference table with all the fields related to attribute.

Field	Detailed Definition
Alert	The count of active, unresolved alerts triggered specifically on this attribute — for example, when a value, threshold, or anomaly detection rule fires for that column. A higher count signals the column needs immediate attention.
Blank Count	The number of rows where this attribute’s value is a blank/empty string (`""`) rather than null. Distinct from Null Count — blanks are present but contain no characters, often indicating upstream data entry issues.
Created At	The timestamp marking when this attribute was first discovered and cataloged by Prizm, typically aligned with when the parent table or column was created at the source.
Criticality	The business-impact classification assigned to this specific attribute (e.g., Critical, High, Medium, Low) — independent of the parent table’s criticality, since individual columns can carry different levels of business sensitivity (e.g., a PII field vs. a metadata timestamp).
Criticality Score	The underlying numeric score (typically 0–100) that determines the Criticality label. Calculated from factors like usage, dependency, and metadata completeness specific to this column.
Data Type	The native data type of the column as defined in the source system (e.g., `NUMBER`, `TEXT`, `BOOLEAN`, `DATE`, `TIMESTAMP_LTZ`). Determines what types of profiling statistics (e.g., Min/Max Value vs. Min/Max Length) are applicable.
Description	The business or technical description written for this attribute, explaining what the column represents and how it should be interpreted or used.
Distinct Count	The number of unique values found in the column across all evaluated rows. Useful for assessing cardinality — low distinct counts on an expected unique key (like an ID) may indicate a data quality problem.
Empty Count	The number of rows where the attribute’s value is empty (no content at all), distinguished from Blank Count and Null Count depending on how the source system differentiates them.
Issue	The count of open, logged data quality issues specifically tied to this attribute — typically created when an alert is escalated or manually flagged for investigation and resolution.
Max Length	For text-type attributes, the maximum character length observed across all values in the column. Helps detect truncation risks or unexpected long-form data.
Max Value	For numeric or date-type attributes, the highest value observed in the column. Useful for spotting outliers or out-of-range data.
Metric Count	The total number of data quality/observability metrics actively configured and tracked on this attribute (e.g., completeness, uniqueness, freshness checks). A higher count generally indicates more thorough monitoring coverage.
Min Length	For text-type attributes, the minimum character length observed across all values in the column. Helps detect unexpectedly short or incomplete entries.
Min Value	For numeric or date-type attributes, the lowest value observed in the column. Useful for spotting negative values, zero anomalies, or out-of-range dates.
Null Count	The number of rows where this attribute’s value is `NULL` (no value recorded at all, as opposed to blank/empty). A key completeness indicator.
Primary Key	A boolean flag indicating whether this attribute serves as (or is part of) the primary key for its parent table — i.e., whether it uniquely identifies each row.
Qualified Name	The full hierarchical path that uniquely identifies this attribute within the platform, typically structured as `Database / Schema / Table / Column`.
Row Count	The total number of rows evaluated when calculating this attribute’s statistics — generally matches the parent table’s row count at time of profiling.
Score	The overall data quality score for this specific attribute, expressed as a percentage, derived from the combined results of all metrics and tests configured on it.
Space Count	The number of values in the column that consist only of whitespace characters (spaces, tabs) rather than meaningful content — often an indicator of dirty or improperly cleaned source data.
Status	The workflow/review status of the attribute within Prizm’s governance process (e.g., `Pending`, `Ready for Review`, `Approved`), reflecting whether it has been reviewed and documented.
Test Count	The number of explicit data quality tests (e.g., not-null, uniqueness, accepted-values) configured specifically on this attribute, separate from passive observability metrics.
Zero Values	A boolean/indicator flag showing whether zero values are present in this numeric column at all.
Zero Values Count	The exact count of rows where this numeric attribute’s value equals zero — useful for distinguishing legitimate zero entries from missing or default-filled data.