DAA-C01 Question 66
Single answerData lineageA data analyst needs to investigate why a dashboard metric changed after a recent release. The metric is sourced from a curated table in Snowflake, and the analyst must identify which upstream objects contributed to the current table values and assess the impact if a source object changes. The organization wants to use built-in Snowflake capabilities rather than maintaining custom metadata tables. Which approach should the analyst use?
- A
Query Snowflake lineage information to trace upstream and downstream object dependencies for the curated table.
- B
Use only QUERY_HISTORY to manually read SQL text and infer all upstream dependencies for the table.
- C
Review TAG assignments on the curated table, because tags automatically provide full transformation-level lineage across all upstream objects.
- D
Use ACCESS_HISTORY alone, because it is designed to show complete object lineage and impact analysis without any lineage feature.
Show answer and explanation
Correct answer: A
Explanation
The best answer is to use Snowflake's built-in lineage capabilities to trace upstream and downstream dependencies for the curated table. In a real-world incident such as a changed dashboard metric, lineage helps analysts determine where data originated, how it moved through transformations, and which downstream assets may be affected by changes. This is more efficient and reliable than manually reconstructing dependencies from query text. QUERY_HISTORY and ACCESS_HISTORY can support investigations, but they are not substitutes for dedicated lineage metadata. Tags are valuable for governance and classification but do not model data flow. This aligns with Snowflake best practices for data governance and impact analysis, where lineage is used to understand dependencies between objects and support change management, troubleshooting, and trust in analytical outputs.
- A. Correct.
Correct. Snowflake provides built-in lineage capabilities that help users trace relationships between data objects and understand upstream and downstream dependencies. This is the most appropriate approach when the goal is to identify contributing source objects and evaluate impact using Snowflake-managed metadata rather than custom solutions. For a curated table used in analytics, lineage is specifically intended to support root-cause analysis, impact assessment, and governance workflows.
- B. Incorrect.
Incorrect. QUERY_HISTORY can sometimes help investigators inspect executed SQL, but it is not the best built-in mechanism for complete lineage analysis. It requires manual parsing of statements, can be time-consuming, and does not provide the purpose-built upstream/downstream lineage view expected for impact analysis. This option reflects a common misconception that query text history alone is equivalent to lineage metadata.
- C. Incorrect.
Incorrect. Tags are useful for governance, classification, and policy-driven management, but they do not automatically provide full object-to-object or transformation-level lineage. A candidate might choose this option because tags are metadata, but tags answer questions such as sensitivity or ownership classification, not the flow of data through transformations.
- D. Incorrect.
Incorrect. ACCESS_HISTORY is useful for understanding which objects were accessed by queries and can support auditing and investigation, but by itself it is not the primary built-in feature for complete lineage and impact analysis. It can complement lineage analysis, but it does not replace Snowflake's dedicated lineage capabilities for tracing data flow across related objects.