Skip to main content

Dataset Catalog

The Dataset Catalog lets you browse, inspect, and manage all datasets in your workspace.

Viewing Your Datasets

Navigate to Datasets in the sidebar to see all available datasets.

Screenshot of the datasets list with table names, descriptions, and action buttons

Each dataset shows:

  • Name — the table name in the data warehouse (e.g., sales_data, customer_list)
  • Description — what the dataset contains
  • Actions — View details or Delete

System datasets (created by the platform for KPIs and dashboards) are protected and cannot be deleted.

Dataset Detail View

Click View on any dataset to see its full details:

Screenshot of a dataset detail page showing the schema viewer with column names, types, descriptions, and a data preview table

Schema Viewer

The schema table shows every column with:

FieldDescription
Column NameThe column identifier
TypeData type (string, number, date, boolean, etc.)
DescriptionHuman-readable description of what the column contains
PIIWhether the column contains personally identifiable information
KeyWhether the column is a primary/business key

Data Preview

A sample of rows from the dataset is displayed in a scrollable table so you can verify the data looks correct.

Editing Metadata

Click the Edit toggle to modify:

  1. Dataset description — Update what the dataset contains
  2. Column descriptions — Add or change descriptions for each column
  3. PII flags — Mark/unmark columns as containing sensitive data
  4. Key flags — Designate primary/business key columns

Click Save when done. Updated metadata immediately improves AI accuracy when answering questions about this dataset.

warning

Important: Good column descriptions dramatically improve AI query accuracy. Instead of just "amount", describe it as "Net sale amount in BRL after discounts, excluding taxes".

Linked KPIs

Some datasets are created by the KPI system (e.g., materialized KPI tables). These show a Linked KPIs section listing which KPIs depend on this dataset.

Deleting a Dataset

  1. Click the Delete button on the dataset row
  2. Confirm in the dialog

Deleted datasets are permanently removed from the data warehouse. This action cannot be undone.