Question 120
Acceptance test: A Product dimension contains duplicates because it stores one row per product color variant while the fact is keyed by base ProductID. The analyst used the following solution: Align the lookup's key to the model's intended unique grain or introduce a ProductVariant key as needed. Which evidence should the reviewer expect?
Explanation
Correct Answer
This checks the actual result rather than a saved setting. Fact and dimension grains must agree on what uniquely identifies a dimension row. A failed implementation may show: Forcing a one-to-many relationship with duplicate supposed dimension keys creates invalid or ambiguous joins.
Incorrect Answers
A customer buying in two months counts once in the overall two-month distinct customer total. However, this does not establish the required outcome: the one-side dimension key is unique and the fact-to-dimension join is stable.
The prior-year value for March 2026 aligns to March 2025 under normal calendar rules. However, this does not establish the required outcome: the one-side dimension key is unique and the fact-to-dimension join is stable.
FY 2027 consistently spans July 2026 through June 2027 as specified. However, this does not establish the required outcome: the one-side dimension key is unique and the fact-to-dimension join is stable.