Data classification and tagging in Snowflake - Commands & Configuration

Introduction

When you have many data tables, it can be hard to know what kind of data each table holds. Data classification and tagging help you label your data so you can find and protect it easily.

When you want to mark sensitive data like personal information to control who can see it

When you need to organize data by categories such as finance, marketing, or operations

When you want to quickly search for data tables based on their tags

When you need to comply with rules that require labeling data for privacy or security

When you want to track data usage by tagging data with ownership or purpose

Commands

This command creates a tag named 'sensitive_data' with a description explaining its purpose.

Terminal

CREATE TAG sensitive_data COMMENT = 'Marks data that contains sensitive information';

Expected Output

Tag SENSITIVE_DATA successfully created.

This command applies the 'sensitive_data' tag to the 'sales_data' table, marking it as sensitive.

Terminal

ALTER TABLE sales_data SET TAG sensitive_data = 'true';

Expected Output

Table SALES_DATA altered.

This command lists all tags matching the name 'sensitive_data' to verify the tag exists.

Terminal

SHOW TAGS LIKE 'sensitive_data';

Expected Output

created_on           | name           | database_name | schema_name | comment

---------------------|----------------|---------------|-------------|-----------------------------

2024-06-01 12:00:00  | SENSITIVE_DATA | MY_DATABASE   | PUBLIC      | Marks data that contains sensitive information

This command shows all tables that have the 'sensitive_data' tag applied, helping you find sensitive data quickly.

Terminal

SELECT TABLE_NAME, TAG_VALUE FROM INFORMATION_SCHEMA.TAG_REFERENCES WHERE TAG_NAME = 'SENSITIVE_DATA';

Expected Output

TABLE_NAME | TAG_VALUE

-----------|----------

sales_data | true

Key Concept

If you remember nothing else from this pattern, remember: tagging data lets you label and find important data easily for security and organization.

Common Mistakes

Trying to apply a tag before creating it

Snowflake will return an error because the tag does not exist yet.

Always create the tag first using CREATE TAG before applying it to tables.

Using lowercase tag names in queries after creating tags with uppercase names

Snowflake treats tag names as case-insensitive but stores them uppercase, causing confusion in queries.

Use uppercase tag names in queries or double quotes to match exact case.

Not checking which tables have tags applied

You might miss important data that needs protection or organization.

Use INFORMATION_SCHEMA.TAG_REFERENCES to list tagged tables regularly.

Summary

Create tags with CREATE TAG to label data meaningfully.

Apply tags to tables with ALTER TABLE SET TAG to mark data.

Use SHOW TAGS and INFORMATION_SCHEMA.TAG_REFERENCES to verify and find tagged data.