ADA-C01 Question 185
Single answerIdentify tagging use casesA healthcare company stores patient and billing data in Snowflake across multiple schemas and wants a scalable way to identify sensitive assets so governance teams can classify data and drive downstream masking and reporting policies. The administrators want a solution that works at the object and column level, is easy to query centrally, and can be applied consistently as new objects are created. Which Snowflake capability is the BEST fit for this requirement?
- A
Create tags such as DATA_SENSITIVITY and DATA_OWNER, assign them to databases, schemas, tables, and columns, and query tag references to inventory classified assets
- B
Create row access policies on all sensitive tables so the policy definitions serve as the central catalog of which assets contain regulated data
- C
Use virtual warehouse names to indicate sensitivity levels, and require developers to run sensitive workloads only on specifically named warehouses
- D
Create separate Snowflake users for each data classification level and infer which objects are sensitive based on object ownership
Show answer and explanation
Correct answer: A
Explanation
The best answer is to use Snowflake tags. A core tagging use case is attaching governance metadata such as sensitivity, privacy class, owner, cost center, or retention category to Snowflake objects and columns. This allows administrators and governance teams to identify and inventory regulated data consistently across the environment. In practice, tags are preferable here because they separate classification metadata from access enforcement logic. Snowflake also supports querying tag assignments through account usage and information schema metadata, which makes centralized reporting practical. In addition, tags integrate well with governance workflows, including tag-based masking strategies in supported scenarios. By contrast, row access policies and ownership are access-control mechanisms, not classification catalogs, and warehouse names are unrelated to object-level data classification. This aligns with Snowflake best practices for using tags to classify and manage governed data assets.
- A. Correct.
Correct. Tags are designed for classification and governance metadata in Snowflake. They can be applied to supported object types, including columns, making them well suited for identifying sensitive data and data ownership. Tags can then be queried centrally using Snowflake metadata views/functions for governance reporting and can support downstream governance patterns such as masking based on tag values.
- B. Incorrect.
Incorrect. Row access policies control which rows a role can see, but they are not intended to serve as a metadata classification system for identifying sensitive assets. They also operate at query-time row filtering rather than as reusable business metadata attached broadly across object and column hierarchies.
- C. Incorrect.
Incorrect. Warehouse naming conventions may help with operational organization, but warehouses are compute resources, not data classification metadata containers. Naming warehouses does not identify which tables or columns are sensitive and does not provide object- or column-level governance metadata.
- D. Incorrect.
Incorrect. Object ownership reflects administrative control, not data sensitivity classification. Creating users by classification level would be difficult to manage, would not identify sensitive columns, and would misuse Snowflake access control concepts for a metadata-governance use case.