PL-300 Prepare the data Practice Question
Which TWO actions should you take to reduce the size of a Power BI dataset? (Choose two.)
⚠ Common exam trap
It's easy for candidates to confuse 'improving performance' with 'reducing dataset size' — disabling query folding can hurt performance and does not reduce size, while DirectQuery changes the architecture rather than reducing an existing Import dataset's size.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Filter out rows that are not needed.
Option A is correct because filtering out rows that are not needed during import (for example, using Power Query filters or a WHERE clause in the source query) reduces the number of records loaded into the model, directly shrinking the dataset size in memory and on disk. Option B is correct because removing unnecessary columns during import eliminates entire columns of data from the model, which reduces both row-level storage and the columnar compression footprint in the VertiPaq engine. Option C is incorrect because disabling query folding typically hurts performance and does not reduce dataset size; folding pushes transformations back to the source. Option D is incorrect because adding calculated columns increases the model's size by storing additional materialized values. Option E is incorrect because DirectQuery does not store data in the model at all, but it is a connectivity mode change rather than a size-reduction action, and it does not reduce the size of an existing Import-mode dataset.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
Filter out rows that are not needed.
Why this is correct
Filtering out rows that are not needed reduces the volume of data loaded into the Power BI model, directly lowering the number of values that VertiPaq stores and compresses. The VertiPaq engine compresses data column-by-column, so fewer rows mean smaller dictionaries and bitmaps, which can also improve compression ratios. However, filter pushdown via query folding may further improve refresh speed, but the size reduction is the primary benefit.
- ✓
Remove unnecessary columns during import.
Why this is correct
Removing unnecessary columns during import eliminates entire columnar structures—dictionaries, bitmaps, and data segments—that would otherwise be stored even if the column is never used. Since Power BI's columnar storage allocates memory per column, each unused column adds a fixed cost regardless of row count. This is often the most impactful way to reduce dataset size because a column can consume significant memory even with few rows, and removing it before any transformations avoids carrying it through the pipeline.
- ✗
Disable query folding to improve performance.
Why it's wrong here
Disabling query folding to improve performance is incorrect because query folding affects how data is transformed at the source versus in Power Query, not the final dataset size stored in the model. Query folding pushes transformations to the source database, which can reduce refresh time and temporary memory usage, but the compressed model still contains the exact same rows and columns after refresh. Disabling it would force transformations into Power Query, potentially slowing refresh without changing the stored dataset's footprint.
- ✗
Add calculated columns to precompute values.
Why it's wrong here
Adding calculated columns to precompute values increases dataset size because each calculated column is stored as a new physical column with its own dictionaries and compression structures in VertiPaq. While calculated columns may improve query performance by precomputing values, they consume additional memory both during storage and when processing queries. Any column added to the model—whether imported or calculated—expands the columnar storage footprint, making the dataset larger rather than smaller.
- ✗
Use DirectQuery instead of Import.
Why it's wrong here
Using DirectQuery instead of Import completely avoids storing data in Power BI, so it does not reduce an existing dataset's size—it eliminates the imported model altogether. DirectQuery stores only metadata and queries the source at report time, moving memory usage to the source system and adding latency. This option is fundamentally different from size reduction: it addresses architecture choice rather than shrinking the imported dataset, and it introduces performance trade-offs that are not part of the stated requirement.
Go deeper
Related to this question
About these practice questions
This PL-300 question is part of Courseiva's 524-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
This PL-300 practice question is part of Courseiva's free Microsoft certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the PL-300 exam.