Tableau Collected Data

This page lists what Sifflet imports from Tableau once you connect it. Tableau Cloud and Tableau Server import the same assets and metadata.

Imported Assets

Sifflet imports the following assets from the Tableau site, through the Tableau Metadata API:

Dashboards
Sheets (worksheets), with the query that feeds each sheet
Metrics (legacy Tableau metrics), when your site has any
Data sources embedded in workbooks
Database tables that these data sources read, shown with the Table type in Sifflet

Workbooks and projects aren't imported as separate assets: Sifflet adds the workbook and project names to the name of each dashboard, sheet, and data source.

The Sifflet Insights browser extension supports the following Tableau asset types:

Views

Collected Metadata

Asset type	Metadata
Dashboard	Name, workbook and project names, sheets it contains, usage (number of views over the last 14 days)
Sheet	Name, workbook and project names, data sources it uses
Metric	Name, underlying view, data sources it uses
Data source	Name, workbook and project names, database tables it reads
Database table	Full name, database, and database connection (connection type, host, port, database, schema)

Sifflet reads dashboard usage from the Tableau usage statistics, and all other metadata from the Metadata API.

Lineage

Sifflet builds Tableau lineage from the Metadata API:

Within Tableau: database tables → data sources → sheets → dashboards, and sheets or dashboards → metrics.
To your data warehouse: Sifflet reads the database connections used by your Tableau data sources (connection type, host, database) and links each database table to the matching table in your Sifflet sources. This works for connections to the data platforms and databases that Sifflet supports, such as Snowflake, BigQuery, Redshift, Databricks, Athena, PostgreSQL, MySQL, SQL Server, and Oracle. Add these tables to Sifflet through their own source first.

Data Freshness

Sifflet refreshes Tableau metadata on the source's schedule (weekly by default for BI tools). See Integrations Management for scheduling and manual refreshes.

On Tableau Server, the Metadata API reflects changes once Tableau has indexed them: newly published content can take some time to appear in Sifflet after the next refresh.

Limitations

Lineage depends on what the Metadata API exposes. Content that the token's user can't access is returned with null names and breaks the lineage: see Tableau Troubleshooting.
Database connections of a type that Sifflet doesn't support (files, extracts without a database connection, other databases) are imported without lineage to a Sifflet source.

Data Exposure

Sifflet only reads Tableau metadata (names, structure, connections, view counts). It doesn't read or store the data of your workbooks or data sources. See Security for how Sifflet stores and protects metadata.