Explore / Microsoft Power BI Data Analyst (PL-300) Exam Prep / Practice questions
Microsoft Power BI Data Analyst (PL-300) practice questions
10 questions from the bank behind this course's timed mock. Free to read and to answer, no account needed, and each has its answer and an explanation. They belong to Microsoft Power BI Data Analyst (PL-300) Exam Prep.
Answer them one at a time →Read lesson 1 free
Question 1
You are an analyst booking PL-300. You open the Study guide for Exam PL-300 and see two versions of the skills measured list, plus a change log. Your exam is booked for a date after the newer list takes effect. You want to spend your study time only on content that will be tested. What should you do?
- Study the skills list that applies on your exam date, and use the change log to see which sections changed
- Study the older skills list, because the Practice Assessment questions are taken from it
- Study only the sections that appear in both versions and skip anything that changed
- Use a copy of the skills list saved from a third-party course, because it will not change before your exam
Show the answer
Answer: A. Study the skills list that applies on your exam date, and use the change log to see which sections changed
The guide shows two versions when a change is coming, so the version to study is the one that applies on your exam date, and the change log shows you what moved. Skipping the changed sections is the nearest distractor, but it drops content that will be tested on your date.
Question 2
You are a data analyst at a retailer. A Power Query query named CutoffDate returns only one date value, and three other queries refer to it. Your manager wants CutoffDate to show up and be managed as a parameter, so colleagues can change it from the parameters list without editing M code. You want the least rework. What should you do?
- Create a new parameter in Manage Parameters, delete the CutoffDate query, and then edit the three queries that use it.
- Select Home > Transform data (dropdown) > Edit parameters and add CutoffDate to the list.
- Right-click the CutoffDate query and select Convert to Parameter.
- Create a new parameter whose Suggested Values is set to Query, pointing at CutoffDate.
Show the answer
Answer: C. Right-click the CutoffDate query and select Convert to Parameter.
A query that returns a simple constant, such as a date, can be turned into a parameter in place with Convert to Parameter, so the queries that use it don't need to change. Rebuilding it by hand in Manage Parameters also works, but you'd have to delete the query and fix three queries, which is far more rework.
Question 3
You are building a sales report from a transactions table that has millions of rows. To make refresh faster, a colleague suggests using Group By in Power Query to reduce the table to one row per Region and Month. Report users have said they must be able to drill down to individual transactions. What should you do?
- Use Group By with the Sum operation on Amount, grouped by Region and Month
- Unpivot the Month column so that the table becomes long and narrow
- Keep the transaction rows in the model and calculate the totals with DAX measures
- Use Group By with the Count rows operation so that the transaction count is kept
Show the answer
Answer: C. Keep the transaction rows in the model and calculate the totals with DAX measures
Grouping in Power Query discards the detail rows permanently, so users could no longer drill to individual transactions. You should keep the detail and summarise with DAX measures instead. Count rows still produces only one row per group, so it doesn't keep the transactions either.
Question 4
You're a data analyst on a sales semantic model. The Date table has a text column, Month, with values such as "2017 Aug" and "2018 Jan", covering several years. Report authors say a line chart using Month on the axis doesn't show the months in date order. You need Month to sort chronologically across all years, and you shouldn't change how it displays. What should you do?
- Set Sort by column on Month to the DateKey column (one row per day)
- Change the data type of Month to Date in Power BI Desktop
- Set Sort by column on Month to the MonthNumber column (1–12)
- Set Sort by column on Month to a YearMonth column that holds whole numbers such as 201708
Show the answer
Answer: D. Set Sort by column on Month to a YearMonth column that holds whole numbers such as 201708
A year-month key matches the grain of Month, so each value such as "2017 Aug" gets one key and the keys run in date order. MonthNumber gives the same 8 to every August, so the months from different years get mixed together instead of following the timeline. DateKey has many values for each month, so Power BI refuses the sort.
Question 5
You are a data analyst working on an Import model. The Sales table has an OrderDateKey and a ShipDateKey. Both should relate to the existing 'Date' table. Managers want one slicer for order date and a separate slicer for ship date on the same page, and each slicer must filter on its own. You want to do this in the model, without changing the source. What should you do?
- Add a calculated column to Sales that uses RELATED('Date'[Date]) to bring in the ship date
- Create a calculated table Ship Date = 'Date', then relate Sales[ShipDateKey] to it
- Create a measure that returns the ship date for each order
- Create a calculated table with SUMMARIZECOLUMNS('Date'[Year], "Total Sales", SUM(Sales[Sales Amount]))
Show the answer
Answer: B. Create a calculated table Ship Date = 'Date', then relate Sales[ShipDateKey] to it
A role-playing copy of a dimension is a classic use for a calculated table. A plain table reference gives you a second date table that the ship date slicer can filter on its own, though it costs extra memory. RELATED can only follow a relationship that already exists, so a column built with it can't create the separate ship date filter path the managers need.
Question 6
A stock analyst writes Closing Units = CALCULATE(SUM(Inventory[UnitsOnHand]), LASTNONBLANK('Date'[Date], SUM(Inventory[UnitsOnHand]))). The date table runs to 31 December, but snapshots exist only up to last week. The measure still returns BLANK for the current year. What is the most likely cause, and how do you fix it?
- The date table must not go past the last snapshot. Replace the measure with CLOSINGBALANCEYEAR, which skips future dates.
- The bare SUM in LASTNONBLANK isn't filtered by each date, so every date looks non-blank and 31 December is picked. Replace it with a measure such as [Units], or wrap it in CALCULATE.
- LASTNONBLANK returns the first date with data, not the last. Switch to LASTDATE('Date'[Date]).
- LASTNONBLANK only works on fact table columns. Change the first argument to Inventory[Date].
Show the answer
Answer: B. The bare SUM in LASTNONBLANK isn't filtered by each date, so every date looks non-blank and 31 December is picked. Replace it with a measure such as [Units], or wrap it in CALCULATE.
LASTNONBLANK needs an expression that is re-evaluated for each date. A measure, or an expression wrapped in CALCULATE, does that, but a bare SUM sees the whole year as non-blank. CLOSINGBALANCEYEAR doesn't help, because it uses the calendar's last date, which is the same blank-date problem as LASTDATE.
Question 7
A report designer is building an executive report. Each page has white text boxes on a dark navy area outside the page canvas, set with Wallpaper. Leadership will receive the report as a PDF from Export to PDF. When it's exported, the white text near the page edges can't be read. What's the most likely cause, and what's the best fix?
- Export to PDF ignores the report theme. Download the theme and import it again before you export.
- Export to PDF doesn't include wallpaper. Use the Canvas background with its transparency lowered, so the dark colour sits on the page itself.
- The wallpaper's transparency defaults to 100%. Lower it to 0% so it shows in the PDF.
- The canvas Type is 16:9. Change it to Letter so the PDF keeps the wallpaper.
Show the answer
Answer: B. Export to PDF doesn't include wallpaper. Use the Canvas background with its transparency lowered, so the dark colour sits on the page itself.
Wallpaper is the area outside the page, and Export to PDF leaves it out, so you need a dark Canvas background with its transparency reduced. The transparency distractor mixes up the defaults: wallpaper is already at 0% transparency, and it's the canvas background that starts at 100%.
Question 8
You are a report author. On the Overview page you create a bookmark called "Hide Help" that hides a help overlay group. Every page has a copy of that group with the same name, and every page has a button that runs the "Hide Help" bookmark. When a user clicks the button on the Details page, the overlay is hidden but the user is taken to the Overview page. Users need to stay on the page they are viewing. What should you do?
- Change the button's Action Type from Bookmark to Page navigation.
- Clear the Data option on the "Hide Help" bookmark.
- Set the "Hide Help" bookmark to All visuals instead of Selected visuals.
- Clear the Current page option on the "Hide Help" bookmark.
Show the answer
Answer: D. Clear the Current page option on the "Hide Help" bookmark.
When Current page is cleared, the bookmark applies its settings to whichever page the user is on and doesn't jump back to the page where it was created. Clearing Data only stops filter and slicer state from being restored, so the user would still be sent to the Overview page.
Question 9
A sales director wants one visual to investigate a fall in Sales Amount. She wants to drill through Region, Channel and Product in any order she likes. At each level she wants Power BI to suggest the next split, picking the value that stands out most compared with the other values at that level, not just the biggest one. What should you configure?
- A Decomposition tree using an AI split set to Absolute
- A Key influencers visual, reviewed on the Top segments tab
- A Decomposition tree using an AI split set to Relative
- A column chart, then Analyze > Explain the decrease on the latest data point
Show the answer
Answer: C. A Decomposition tree using an AI split set to Relative
The Decomposition tree is the visual for ad hoc root-cause drilling in any order, and a Relative AI split looks for values that stand out against the other values at the same level. An Absolute split only picks the largest value, so it does not meet the 'stands out' requirement.
Question 10
A finance analyst manages a report whose semantic model is refreshed at 06:00, 10:00, 14:00 and 18:00 every day. She creates an email subscription to the report and sets Repeat to "After data refresh". What will recipients receive?
- One email a day, sent after the last scheduled refresh of that day
- One email a day, sent after the first scheduled refresh of that day
- One email each time someone selects Run now, because Run now triggers a refresh and then sends the email
- Four emails a day, one after each scheduled refresh
Show the answer
Answer: B. One email a day, sent after the first scheduled refresh of that day
The "After data refresh" option sends at most once a day, after the first scheduled refresh of that day, so later refreshes don't send more emails. Run now sends the email straight away but never triggers a data refresh.
The full timed mock
50 questions in 100 minutes, with a pass mark of 70%. Each paper is drawn from a bank of questions written for the mock alone, and these 10 are left out of it. Full access opens it, with every level of the course.
LevelUpWith is independent of the organisation that sets this exam. These questions were written for this course; they are not taken from the exam.