PrepFree PL-300 practice test
20 original questions written by our editors against the official Microsoft exam domains. They are not real exam questions. Pick an answer to see why it is right or wrong.
-
Question 1 of 20Prepare the data
A logistics company has a 2-billion-row shipments table in an on-premises SQL Server. Managers need report visuals to reflect changes within a minute, and the table is too large to import into the semantic model. Which storage mode should the analyst use for this table?
Show answer
Answer: C. DirectQuery leaves the data in the source and sends queries to it when visuals load, so results stay close to real time and the table does not need to fit in the model. Import and incremental refresh only show data as fresh as the last refresh. Dual still caches an imported copy that depends on refresh.
-
Question 2 of 20Prepare the data
An analyst has a Power Query query called Customers with ten cleaning steps. She needs a second query that starts from the cleaned Customers result and always picks up any future changes to those ten steps. What should she create?
Show answer
Answer: D. A reference creates a new query whose source is the output of the original query, so later edits to Customers flow through. A duplicate or a pasted copy is an independent copy of the steps that will not follow future changes. A self-merge joins rows and is not a way to branch a query.
-
Question 3 of 20Prepare the data
A budget file has one row per cost centre and separate columns for Jan, Feb, Mar and so on. New month columns are added during the year. The analyst needs a table with Cost Centre, Month and Amount columns that keeps working when new months appear. What is the best transformation?
Show answer
Answer: A. Unpivot Other Columns keeps the selected column and unpivots everything else, so new month columns are included automatically. Unpivoting a named list of month columns hard-codes those columns and misses new ones. Transpose swaps all rows and columns, and pivoting goes in the opposite direction.
-
Question 4 of 20Prepare the data
In Power Query Editor, the column distribution shows that an Order ID column has no duplicates, yet the model later reports duplicate keys. The source table has 900,000 rows. What is the most likely reason the profile looked clean?
Show answer
Answer: B. By default the column quality, distribution and profile tools evaluate the top 1,000 rows, so duplicates further down are missed. The setting can be switched to profile the entire data set. Profiling works on all column types, and Power Query never removes duplicates unless you add a step to do so.
-
Question 5 of 20Prepare the data
Sales for 2025 and 2026 arrive as two separate tables with identical columns. The analyst wants one Sales table containing all rows from both years. Which Power Query operation should she use?
Show answer
Answer: C. Append stacks the rows of tables with the same structure into one table. Merge joins tables side by side on matching keys, which is used to add columns from a related table. Group by aggregates rows, and a conditional column adds a calculated column.
-
Question 6 of 20Model the data
A Sales table has both OrderDate and ShipDate columns related to the same Date table. Only the OrderDate relationship is active. The analyst needs a measure that totals sales by ship date. What should the measure use?
Show answer
Answer: A. USERELATIONSHIP inside CALCULATE activates the inactive ShipDate relationship for that measure only, which is the standard way to handle a role-playing date dimension. Power BI allows only one active relationship between two tables. RELATED fetches a value from the one side of a row context, and CROSSFILTER changes direction, not which relationship is used.
-
Question 7 of 20Model the data
A Product table has one row per product, and the Sales table has many rows per product. The analyst wants slicers on Product to filter Sales, without Sales filtering Product. How should the relationship be configured?
Show answer
Answer: C. A dimension with unique keys relates to a fact table as one-to-many, and a single cross-filter direction lets the dimension filter the fact table only. Many-to-many and bidirectional filtering add ambiguity and cost. One-to-one is wrong because the Sales key repeats.
-
Question 8 of 20Model the data
A report page has a slicer on Product[Colour]. The analyst needs a measure that always returns sales for blue products, whatever colour the user picks in the slicer. Which DAX expression achieves this?
Show answer
Answer: B. A simple Boolean filter in CALCULATE replaces any existing filter on Product[Colour], so the slicer selection is overridden and blue sales are returned. KEEPFILTERS intersects with the slicer, so selecting red would return blank. The SUMX FILTER version still respects the slicer's filter context, and the SELECTEDVALUE version returns zero unless blue is selected.
-
Question 9 of 20Model the data
A Stock table holds the quantity on hand for each product at the end of every day. A matrix by month currently shows the sum of 30 daily balances, which is meaningless. What kind of measure should replace it?
Show answer
Answer: A. Balances are semi-additive: they can be summed across products but not across time, so the measure should take the value at the last date in the period, for example CALCULATE(SUM(Stock[Qty]), LASTDATE('Date'[Date])). Averages and calculated columns give the wrong figure, and TOTALYTD accumulates values, which is the opposite of what a balance needs.
-
Question 10 of 20Model the data
Time intelligence measures such as year-to-date return unexpected results. The model's date table was generated from the distinct order dates in the Sales table. What should the analyst fix?
Show answer
Answer: D. Time intelligence functions need a date table with contiguous dates covering whole years, marked as the model's date table. A table built from order dates has gaps on days without orders. Text dates break date logic, and bidirectional filters or flattening the model do not solve missing days.
-
Question 11 of 20Model the data
A report page takes 12 seconds to load and the analyst suspects one particular visual. Which built-in tool shows how long each visual's DAX query and rendering take?
Show answer
Answer: B. Performance Analyzer records the duration of the DAX query, visual display and other work for each visual, so slow visuals stand out. Quick insights finds patterns in data, the Selection pane controls visual layering and visibility, and query dependencies shows how Power Query queries connect.
-
Question 12 of 20Visualize and analyze the data
Managers look at a summary page of sales by product category. When they find an unusual category, they want to right-click it and open a detail page that shows only that category's orders. What should the analyst configure?
Show answer
Answer: C. Drillthrough lets a user right-click a data point and jump to a target page that is filtered to that value. Bookmarks save a page state but do not pass the clicked value. A tooltip page appears on hover, and sync slicers align slicer selections across pages.
-
Question 13 of 20Visualize and analyze the data
A report has five pages, each with a Region slicer. Users complain that choosing a region on one page does not carry over to the others. Which feature fixes this with the least effort?
Show answer
Answer: D. Sync slicers keeps the selection of a slicer consistent across the pages you choose. Edit interactions controls how visuals on the same page affect each other, visual-level filters apply to one visual only, and row-level security restricts data by user rather than syncing selections.
-
Question 14 of 20Visualize and analyze the data
On one report page, users want a button to switch between a chart view and a table view of the same data, in the same spot on the page. What is the usual way to build this?
Show answer
Answer: D. Bookmarks can capture which visuals are visible, set through the Selection pane, and buttons can apply a bookmark, creating a toggle. Drillthrough moves to another page with a filter, paginated reports are separate, print-ready documents, and tooltips only appear on hover.
-
Question 15 of 20Visualize and analyze the data
Finance needs monthly supplier statements that print exactly to A4, run to many pages per supplier, and are exported to PDF. Which Power BI option fits best?
Show answer
Answer: D. Paginated reports are designed for pixel-perfect, printable layouts that can span many pages, which suits statements and invoices. A table visual on a report page scrolls instead of paginating cleanly. Dashboards and Q&A are for monitoring and natural-language exploration, not print layouts.
-
Question 16 of 20Visualize and analyze the data
A subscription business wants a visual that ranks which factors, such as contract type or tenure, most increase the likelihood that a customer cancels. Which visual should the analyst use?
Show answer
Answer: B. The key influencers AI visual analyses which factors drive a selected outcome and ranks them. A decomposition tree lets users break a measure down by dimensions manually or with AI splits, but it does not rank drivers of an outcome in the same way. Waterfall charts show running changes, and gauges show progress toward a target.
-
Question 17 of 20Manage and secure Power BI
A semantic model imports data from an on-premises SQL Server and from a SharePoint Online list. The analyst publishes it to the Power BI service and wants a scheduled refresh. What is required?
Show answer
Answer: B. The Power BI service needs an on-premises data gateway to reach data sources inside the corporate network, such as an on-premises SQL Server. Cloud sources such as SharePoint Online can be refreshed directly. Publishing the model to the cloud does not give the service access to on-premises servers.
-
Question 18 of 20Manage and secure Power BI
A colleague must be able to publish, edit and delete reports in a workspace, but must not be able to add people to the workspace or change its permissions. Which workspace role should she get?
Show answer
Answer: C. Contributors can create, edit and delete content in a workspace but cannot add users or share access. Members can also add people with their own role or lower, and Admins have full control, including deleting the workspace. Viewers can only view content.
-
Question 19 of 20Manage and secure Power BI
Each of 40 regional managers should see only their own region's rows in a sales report. A Managers table maps each manager's sign-in email to a region. What is the most maintainable row-level security design?
Show answer
Answer: C. Dynamic RLS uses one role whose filter compares a column to USERPRINCIPALNAME(), so the mapping table controls access and new managers only need a new row. Forty static roles or forty report copies are hard to maintain. Hiding a column does not restrict which rows a user can see.
-
Question 20 of 20Manage and secure Power BI
The analytics team has eight related reports in a workspace and wants to give 500 business users one place to find them, with a navigation menu, while keeping the users away from work-in-progress content. What should the team use?
Show answer
Answer: A. A Power BI app packages selected content from a workspace with its own navigation and audience settings, separating consumers from the workspace itself. Sharing reports one by one is hard to manage, Contributor would let users edit content, and PowerPoint exports are static copies.
| Exam domain | Correct |
|---|