Finding Important Information about Tables and Columns

Estimated reading: 3 minutes

Stemma provides out-of-the-box integration with applications such as Snowflake, dbt, and BI tools such as Looker, Mode and Tableau to capture rich metadata from each of these systems automatically. For example, in the case of Snowflake, Stemma reads both the information_schema and the query logs, and automatically generates important information such as when a table was last updated, column-to-column lineage, table-to-table lineage, table-to-dashboard lineage, frequent queriers of the table, common join and filter conditions, etc.

The Stemma UI makes it easy for you to discover this important information about your tables and columns, and also provides a convenient way to add and modify curated information.

The table details page provides following information:

About

Asset status

This can be Certified, Deprecated, Intermediate, or No Status

Owners

The person or people responsible for maintaining this table

Click on an owner to see other assets owned by that user, as well as their bookmarks and frequently used tables. You can also email the user via this profile page.

Description

A brief summary of the contents of the table

See Advanced Description Editing for more information.

Last Updated

The date and time of the last update to the table (this can be different from the most recent date covered by the data itself)

Date Range

The date range covered by the data in the table.

Columns

Descriptions are automatically ingested from a source (like dbt or Snowflake). If there is no description to ingest, Stemma will automatically use column-level lineage to search upstream for a description in an upstream column with the exact same name. This documentation is shown as “Autodescribed.”

Column descriptions can be created, and autodescribed documentation over-written, within Stemma. Also available in this table are the data type, related glossary terms and badges.

Lineage

dbt Compiled Query

For dbt users, see the query that formed the table, as well as the dbt model type.

Upstream, Downstream

Upstream and downstream tables and dashboards, sorted by popularity.

To see a graphical version of the lineage, check out the “Lineage Graph” button in the upper right. See Displaying Data Lineage in the Stemma UI for more info.

Usage

Frequent Users

People who have frequently used this table, number of queries they’ve run and their most recent queries. Data is shown for the last 90 days.

Click on a user to see assets owned by that user, as well as their bookmarks and other frequently used tables. You can also email the user via this profile page. See also How Does Stemma Calculate “Top 5 Columns”, “Frequent Users”, etc?

Slack Conversations

Slack threads that mention this table

Issues

JIRA issues logged against this table

This section explains how to link a JIRA issue to the Stemma catalog.

Additional Fields

Custom table attributes can also be added via the API. For more info, see Using the External GraphQL API

Curated versus automated metadata

Where possible, Stemma captures the information directly from the source. This table shows which attributes are populated from the source and which you can edit:.

Table Metadata	Automatically captured?	Users can edit?
Table description.	✅	✅
Asset status.	❌	✅
Owners	✅ (dbt)	✅
Tags	✅ (dbt)	✅
Jira issues	❌	✅
Last updated time	✅	❌
Data date range	✅	❌
Commonly used with tables	✅	❌
Frequent users	✅	❌
Upstream & downstream table lineage	✅	❌
Dashboards that use the table	✅	❌
Slack threads with table mentions	✅	❌