PL-300 Prepare the data Practice Questions

Free PL-300 Prepare the data practice test questions

8 questions from this domain with answers and explanations - different from the samples on the main PL-300 page. A free account includes a fixed set of 20 questions for this exam; paid access unlocks the full bank.

1. Question 1Prepare the data

   You need to combine rows from two queries that have the same schema into a single table. Which Power Query operation should you use?

   • AMerge Queries
   • BTranspose
   • CAppend QueriesCorrect
   • DGroup By
   ✓ Correct answer: C

   Append Queries stacks the rows of two or more queries that share the same column schema into a single table, much like a SQL UNION. Because both queries have identical columns, appending simply adds the second query's rows beneath the first query's rows. This is the correct operation when the goal is to combine records rather than to add columns from a related table.

   Why the other options are wrong

   • AMerge Queries joins tables side by side on matching key columns to bring in additional columns, not to stack rows, so it does not combine same-schema records.
   • BTranspose rotates rows into columns and columns into rows within a single table, which does not combine two queries.
   • DGroup By aggregates rows within one query into summarized groups; it does not append rows from a second query.
2. Question 2Prepare the dataSelect all that apply

   You are configuring a Power BI dataflow to prepare data for multiple semantic models. Which three benefits does using a dataflow provide? (Choose three.)

   • AReduced data redundancy across semantic modelsCorrect
   • BReusable preparation logic across semantic modelsCorrect
   • CAutomatic creation of DAX measures for reports
   • DAutomatic row-level security role configuration
   • ECentralized transformation managed apart from reportsCorrect
   ✓ Correct answer: A, B, E

   A Power BI dataflow performs Power Query transformations in the Service and stores the results so multiple semantic models can consume the same prepared entities. This eliminates duplicating the same extract-and-clean logic in every semantic model, reducing redundancy and creating a single, reusable preparation layer. Because the dataflow lives independently of any report, transformation logic is centrally maintained and governed apart from report development.

   Why the other options are wrong

   • CA dataflow stores the output of Power Query transformations in storage that semantic models can consume; DAX measures belong to the semantic model and are written by the modeller, so a dataflow never generates them.
   • DRow-level security is defined on the semantic model as roles with DAX filters, with members assigned in the service; a dataflow shapes and stores data upstream and has no notion of roles or per-user filters.
3. Question 3Prepare the data

   You are connecting Power Query to a REST API that returns JSON data. The response includes a nested record structure with arrays inside objects. Which Power Query function should you use to convert the nested records into tabular format?

   • ATable.ExpandRecordColumnCorrect
   • BTable.FromRecords
   • CJson.Document
   • DRecord.ToTable
   ✓ Correct answer: A

   When a JSON response includes nested record structures (objects containing arrays or other complex types), converting this to tabular format requires expanding those nested records into separate columns. The Table.ExpandRecordColumn function takes a table with record-type columns and expands the fields from each record into individual columns, effectively flattening the nested structure into a two-dimensional table suitable for Power BI analysis.

   Why the other options are wrong

   • BTable.FromRecords creates a table from an existing list of records, not expanding nested structures.
   • CJson.Document parses JSON text into Power Query structures but does not flatten nested records into columns.
   • DRecord.ToTable converts a single record into a two-column table of field names and values, not suitable for tabular expansion of record columns within a table.
4. Question 4Prepare the data

   You need to create a parameter in Power Query to allow users to change the database server name without editing the query. What type of parameter should you create?

   • AA calculated column with the server name
   • BA Power Query parameter of type TextCorrect
   • CA DAX measure with a text value
   • DA bookmark with the server configuration
   ✓ Correct answer: B

   Power Query parameters are designed to allow dynamic values that can be changed without editing the query itself. To parameterize a database server name, you should create a Power Query parameter of type Text containing the server name. This parameter can then be referenced in the database connection string in the M code, allowing users to change the server name through the Power Query parameter dialog in Power BI Desktop or settings in the Power BI service without modifying the actual query logic.

   Why the other options are wrong

   • AA calculated column with the server name is incorrect because calculated columns in Power BI are for creating new data columns in tables, not for parameterizing connection strings; they also cannot be modified at refresh time like parameters.
   • CA DAX measure with a text value is incorrect because DAX measures operate on loaded data and cannot affect connection strings or source configuration.
   • DA bookmark with the server configuration is incorrect because bookmarks are Power BI visualization features for saving visual states, not for managing query parameters or connection configurations.
5. Question 5Prepare the data

   A data analyst needs to create a date column from separate Year, Month, and Day columns in Power Query. Which M function should be used in a custom column?

   • ADate.FromText([Year] & "-" & [Month] & "-" & [Day])
   • BDate.From([Year] & [Month] & [Day])
   • CDateTime.From([Year], [Month], [Day])
   • D#date([Year], [Month], [Day])Correct
   ✓ Correct answer: D

   The #date() function is the correct M language syntax for creating a date value from separate year, month, and day components. The # prefix indicates this is a date literal in M language. The function takes three integer arguments in the format #date(year, month, day) and returns a proper Date type value. This is the standard way to construct dates programmatically in Power Query custom columns when you have separate numeric columns representing these components. The resulting value is a native date type that Power BI recognizes and handles correctly.

   Why the other options are wrong

   • Adate.FromText([Year] & "-" & [Month] & "-" & [Day]) is incorrect because Date.FromText expects a text value in a specific format string, and concatenating numbers without proper padding may produce invalid text.
   • Bdate.From([Year] & [Month] & [Day]) is incorrect because concatenating the numbers creates an invalid integer (e.g., 202362024), not a date value.
   • CdateTime.From([Year], [Month], [Day]) is incorrect because this function does not exist; it's not the correct syntax for M language.
6. Question 6Prepare the data

   An analyst connects to a Dataverse environment that has a table with 2 million rows. Power Query shows that the applied steps include a filter and a column selection. The analyst notices the refresh is very slow. What should the analyst check first?

   • ARemove all applied steps and reimport the raw Dataverse table, then reapply the filter and column selection afterward
   • BVerify that query folding is occurring by right-clicking steps and checking if 'View Native Query' is availableCorrect
   • CSwitch to importing from a manually exported CSV copy of the Dataverse table to bypass the connector entirely
   • DIncrease the Power BI Desktop maximum memory allocation in the Options dialog to make the refresh run faster
   ✓ Correct answer: B

   When a Dataverse table has millions of rows, the performance of your query depends critically on query folding. If folding is not occurring, all 2 million rows are loaded locally and then filtered and filtered in Power Query, which is very slow. You can verify folding status by right-clicking each applied step: if 'View Native Query' is available as an option, that step is foldable and folding is occurring. If this option is grayed out or unavailable, that step breaks folding. This diagnostic step should be performed first before trying solutions that might remove necessary transformations.

   Why the other options are wrong

   • ARemoving and re-adding steps recreates the same query without diagnosing why folding is not occurring.
   • CA static CSV export loses the live connection and refresh capability and is not a performance diagnosis.
   • DDesktop memory settings do not address a slow source query; the issue is whether the steps fold to Dataverse.
7. Question 7Prepare the dataSelect all that apply

   An administrator at Tailwind Traders is planning to use column quality. Which two of the following are requirements or features of this solution? (Choose two.)

   • APrivacy levelsCorrect
   • BQuality column
   • CError handling in queries
   • DColumn distribution
   • EM languageCorrect
   ✓ Correct answer: A, E

   Column quality features require consideration of privacy levels to ensure that quality monitoring and data profiling do not expose sensitive information through analysis results or diagnostic data. M language support is fundamental because implementing custom quality rules, exception handling, and quality scoring logic often requires writing M expressions to define what constitutes acceptable data quality within your specific business context.

   Why the other options are wrong

   • BQuality column is incorrect because while column quality creates quality indicators, it is not a separate requirement.
   • CError handling in queries is incorrect because error handling and column quality monitoring are separate concerns, though both can coexist in a solution.
   • DColumn distribution is incorrect because distribution analysis is a quality assessment tool rather than a requirement of the quality feature itself.
8. Question 8Prepare the data

   A query reads from a SQL Server stored procedure that returns sales data. You must pass a parameterized native query and ensure Power BI does NOT attempt to wrap or fold additional steps into a query it cannot interpret. Which approach is MOST appropriate?

   • AUse Value.NativeQuery against the Sql.Database connection, supplying the statement and parameters, with EnableFolding set as neededCorrect
   • BType the SQL into the 'SQL statement' box of the Sql.Database navigator and rely on automatic folding to optimize the parameters
   • CImport the entire source table with Sql.Database and filter down to the stored procedure's output afterwards in Power Query
   • DWrap the stored procedure call in Table.Buffer so that the mashup engine will treat the result as foldable to the server
   ✓ Correct answer: A

   Value.NativeQuery(target, query, parameters, options) sends a raw native statement to the source and binds parameters, which is the supported pattern for calling stored procedures or passing custom SQL. You can include the EnableFolding option in the options record to influence whether subsequent steps fold on top of the native query. This avoids the engine guessing how to translate a stored procedure and gives you explicit control over the executed statement. Hand-authored SQL in the navigator works but does not parameterize cleanly and the engine generally treats it as a non-folding boundary.

   Why the other options are wrong

   • BThe SQL statement box on the connector dialog runs a native statement, but it concatenates any values into the text rather than binding them as parameters, and steps placed after a stored procedure call generally cannot fold, so automatic folding will not optimise it.
   • CImporting the whole table transfers every row across the network and then filters locally, defeating the point of executing the procedure on the server, and Power Query cannot reproduce a stored procedure's logic by filtering its source table.
   • DTable.Buffer reads a table fully into the mashup engine's memory and explicitly ends query folding from that point onward; it is used to stabilise a result and can never make a step foldable.

How Prepare the data is tested

This domain holds 197 of the 706 questions in the PL-300 bank, about 28%. The mix is 112 single-answer multiple choice, 39 multiple-response, 28 true/false, 8 yes/no scenario, 5 hotspot and 5 ordering, so it is worth practising the formats as well as the content.

Once you have a few attempts recorded, CertGrid scores every domain separately and points you at the weakest one, so you can drill Prepare the data on its own rather than re-running full-length mocks.

Other PL-300 exam domains

• Model the data188 questions
• Visualize and analyze the data186 questions
• Manage and secure Power BI135 questions
• All PL-300 practice questions706 total
• Prepare the data study notesKey concepts
• Microsoft practice examsAll Microsoft

PL-300 Prepare the data FAQ