Which DAX pattern optimizes a measure by using a variable and SWITCH for the result?

Which DAX pattern optimizes a measure by using a variable and SWITCH for the result?

• A. SWITCH( TRUE(), [Variance] > 0.80, "Amazing!", [Variance] > 0.60, "Good", "Bad" )
• B. VAR Calc = [Variance] RETURN SWITCH(TRUE(), Calc > 0.80, "Amazing!", Calc > 0.60, "Good", "Bad" )
• C. VAR Calc = [Variance] RETURN SWITCH( TRUE(), Calc > 0.80, "Amazing!", [Variance] > 0.60, "Good", "Bad" )
• D. VAR Calc = [Variance] RETURN SWITCH( TRUE(), [Variance] > 0.80, "Amazing!", [Variance] > 0.60, "Good", "Bad" )

Answer: (B)

Caching the measure result in a variable and using SWITCH with TRUE to classify into labels is the pattern that optimizes DAX. By computing the measure once and reusing that single value in every condition, the engine avoids recomputing the measure for each test and ensures the same value drives all comparisons. This improves performance and makes the logic clearer. The best answer defines a variable to hold the measure (Calc = [Variance]) and then uses that variable in all the SWITCH tests (Calc > 0.80, Calc > 0.60). This ensures the measure is evaluated once per evaluation context and the branches rely on the same value, producing consistent results. Other patterns either compute the measure directly in each condition or mix direct references with the variable. That leads to repeated evaluations of the measure or inconsistent usage of the stored value, which is less efficient and harder to maintain.

Caching the measure result in a variable and using SWITCH with TRUE to classify into labels is the pattern that optimizes DAX. By computing the measure once and reusing that single value in every condition, the engine avoids recomputing the measure for each test and ensures the same value drives all comparisons. This improves performance and makes the logic clearer.

The best answer defines a variable to hold the measure (Calc = [Variance]) and then uses that variable in all the SWITCH tests (Calc > 0.80, Calc > 0.60). This ensures the measure is evaluated once per evaluation context and the branches rely on the same value, producing consistent results.

Other patterns either compute the measure directly in each condition or mix direct references with the variable. That leads to repeated evaluations of the measure or inconsistent usage of the stored value, which is less efficient and harder to maintain.