Replacing the FILTER function with KEEPFILTERS reduces execution time in which scenario?

• A. FILTER uses a nested CALCULATE
• B. FILTER references a measure
• C. FILTER references columns from multiple tables
• D. FILTER references a column from a single table that uses Import mode

Answer: (D)

When you use FILTER, the engine builds a new table by evaluating the condition across the chosen table. That often means additional table materialization and a more complex filter-context propagation, which adds overhead to the calculation.

Keeping the existing filter context with KEEPFILTERS avoids that extra table creation step and simply augments the current filters. In the specific case of filtering a column from a single table that uses Import mode, the data is already loaded in memory, so avoiding the creation of a separate filtered table and the extra context transitions yields a noticeable speedup.

If the filter involves nested calculations, a measure, or filters multiple tables, the performance gain from using KEEPFILTERS isn’t as predictable because the engine has to handle more complex filter propagation or cross-table interactions.