Statistics

Column-level statistical profiling computed per snapshot — surfacing distributions, null counts, and value ranges to help assess data quality.

Dataset Statistics provides a statistical profile of the data in a Dataset. Statistics are computed at the column level, surfacing distribution information, null counts, distinct value counts, and value ranges — helping you assess data quality and understand content before applying transformations.

Statistics can be calculated automatically during ingestion by enabling the Calculate Stats option in the Discovery step, or reviewed at any time from the dataset view.

Accessing Statistics

1. Navigate to Data Studio > Dataset.

2. Locate the Dataset within its Workspace and Collection and open it.

3. Click Statistics in the Quick Links section at the top right of the dataset view.

The Statistics panel opens, showing per-column profiling data for the selected Dataset.

Time Range and Snapshot Navigation

Statistics are computed per snapshot instance. At the top of the panel, select the time window to view:

Option	Description
Today	Statistics from today's snapshot.
Current Week	Statistics from the current calendar week (default).
Week	A rolling 7-day window.
Bi Monthly	A rolling two-month window.
Month	A rolling 30-day window.
Year	A rolling 12-month window.
All	All available historical snapshots.

A horizontal instance timeline displays each available snapshot as a point. Navigate between instances using the ◀ and ▶ arrows. The active snapshot is highlighted in blue with its timestamp displayed (e.g., 04/28/2026 14:29:51). The Instance Range label shows the full date span covered by the selected window.

Anomaly Indicators

Each snapshot point on the timeline is colour-coded to reflect its data quality status:

Colour	Status	Meaning
🔵 Blue	Non-Anomalous	The snapshot is within expected statistical bounds.
🔴 Red	Anomalous	One or more columns have values outside expected bounds — potential data quality issue.
🟡 Yellow	Pending Decision	The snapshot has been flagged but not yet reviewed or classified.

What Statistics Show

For each column in the Dataset, the statistics table provides the following metrics:

Metric	Description
Stats Info	The column name.
Total Count	The total number of rows in the snapshot.
Approx Distinct Count	The approximate number of unique values in the column.
Blank Count	The number of rows where the column contains an empty string (NA if not applicable for the data type).
Null Count	The number of rows where the column value is null or missing.
Min Value	The smallest value in the column for this snapshot. Shown as a date/time for temporal columns; NA for string columns.
Max Value	The largest value in the column for this snapshot.

Example:

Stats Info	Total Count	Approx Distinct Count	Blank Count	Null Count	Min Value	Max Value
CREATED_DATE	5	2	NA	3	2024-10-29 06:28:41	2025-08-14 07:...
END_TIME	5	5	NA	0	2026-04-27 04:51:51	2026-04-28 07:...
ACCOUNT_UUID	5	0	0	5	NA	NA
EMAIL_ADDRESS	5	5	0	0	NA	NA
ORGANIZATION	5	2	0	0	NA	NA
SESSION_ID	5	5	0	0	NA	NA
USER_NAME	5	5	0	0	NA	NA

Inline Distribution Charts

Column-level distribution charts are also displayed inline above each column header in the Dataset Table view, giving a quick visual overview of value frequency without navigating to the Statistics page. These histograms update to reflect the current dataset version.

When Statistics Are Computed

Statistics can be generated in during Dataset creation or Dataset write. Statistics are computed automatically as part of the submitted job.