This post is a part of the PL-300: Microsoft Power BI Data Analyst Exam Prep Hub; and this topic falls under these sections:
Model the data (25–30%)
--> Optimize model performance
--> Improve Performance by Reducing Granularity
Below are 10 practice questions (with answers and explanations) for this topic of the exam.
There are also 2 practice tests for the PL-300 exam with 60 questions each (with answers) available on the hub.
Practice Questions
1. A Power BI model contains a Sales table with one row per transaction, including TransactionID, DateTime, ProductID, and SalesAmount. Reports only show daily sales totals. What is the BEST way to improve performance?
A. Create a calculated column to extract the date
B. Create a measure that sums SalesAmount
C. Aggregate the Sales table to daily totals in Power Query
D. Create a calculated table using SUMMARIZE
Correct Answer: C
Explanation:
Aggregating the table to daily totals before loading the data reduces row count and model size. Power Query aggregation is more efficient than DAX-based aggregation and is a best practice for performance optimization.
2. Which type of column most commonly increases granularity without adding analytical value?
A. Date
B. Product category
C. Transaction ID
D. Sales amount
Correct Answer: C
Explanation:
Transaction IDs are typically unique and dramatically increase granularity and cardinality. If they are not used for analysis, they should be removed to improve performance.
3. A dataset refresh is slow due to millions of rows from an IoT table with second-level timestamps. Reports only analyze data by day. What should you do?
A. Convert the timestamp to text
B. Remove the timestamp column
C. Aggregate the data by date in Power Query
D. Use DirectQuery instead of Import
Correct Answer: C
Explanation:
Aggregating the data by date significantly reduces granularity and row count, improving both refresh and query performance.
4. What is the PRIMARY trade-off when reducing granularity in a fact table?
A. Increased memory usage
B. Reduced model refresh frequency
C. Loss of detailed drill-down capability
D. Slower DAX calculations
Correct Answer: C
Explanation:
Reducing granularity improves performance but can limit the ability to drill into detailed records, such as individual transactions.
5. Which approach is MOST appropriate when users need both summary-level performance and occasional transaction-level analysis?
A. Keep only transaction-level data
B. Reduce granularity and remove detail
C. Use aggregation tables with a detailed fact table
D. Create calculated measures for aggregation
Correct Answer: C
Explanation:
Aggregation tables allow Power BI to use summarized data for most queries while retaining access to detailed data when needed.
6. Why does reducing granularity improve VertiPaq engine performance?
A. It increases the number of relationships
B. It reduces the number of visuals per report
C. It lowers row count and column cardinality
D. It forces single-direction filtering
Correct Answer: C
Explanation:
VertiPaq performs best with fewer rows and lower cardinality. Reducing granularity directly improves compression and query speed.
7. Which of the following is the BEST indicator that a table’s granularity is too high?
A. Many measures use CALCULATE
B. The table has a high row count but few visuals use it directly
C. The table contains many numeric columns
D. The table uses Import mode
Correct Answer: B
Explanation:
If a table contains massive detail that is never used in visuals or analysis, its granularity is likely higher than required.
8. A fact table includes a DateTime column with values down to the second. Reports only use Year and Month. What is the BEST action?
A. Hide the DateTime column
B. Replace DateTime with a Date column
C. Convert DateTime to text
D. Create a Year-Month calculated column
Correct Answer: B
Explanation:
Replacing DateTime with a Date column reduces cardinality and improves performance. Simply hiding the column does not improve the model.
9. Which optimization should be done FIRST when addressing performance issues caused by excessive granularity?
A. Rewrite DAX measures
B. Change relationship direction
C. Reduce rows and columns during data preparation
D. Enable bidirectional filtering
Correct Answer: C
Explanation:
Model-level optimizations—such as removing unnecessary rows and reducing granularity—should always be done before tuning DAX.
10. Which statement BEST reflects a PL-300 best practice regarding granularity?
A. Always store the most detailed data possible
B. Reduce granularity only after publishing the report
C. Match data granularity to actual reporting requirements
D. Use calculated tables instead of Power Query aggregation
Correct Answer: C
Explanation:
The optimal granularity is the lowest level of detail that still supports business questions. Overly detailed data harms performance without adding value.
Go back to the PL-300 Exam Prep Hub main page
The Data Community 