DP-700 Sample Questions With Answers

About these questions

These five questions are original. We wrote them to match the style of the DP-700 skills measured, covering all three domains. They are not real exam questions, and the real exam will word things differently. You can read how we write practice questions on our how we write our questions page.

How to use these

Try each question without looking at your notes, and pick an answer before you open the explanation. Then read the explanation even when you were right, because the reason the closest wrong option fails is where the learning is. Note any domain where you slipped and revisit it in the domains explained page.

Question 1 (Implement and manage an analytics solution)

A team uses a deployment pipeline with Development, Test, and Production stages. Someone fixed a bug directly in a notebook in Production. The team wants to copy just that notebook back to Test, which already contains other items. What is the result?

  • A. The notebook can be selected and deployed backward on its own
  • B. A deployment rule for Test makes the backward deployment work
  • C. It is not possible, because backward deployment needs an empty target stage and deploys all items
  • D. It works if the user has Contributor access in both workspaces
Show the answer and explanation

Correct answer: C. Deploying backward, from a later stage to an earlier one, is only possible when the earlier stage is empty, and it deploys everything rather than selected items. Test already has content, so the team cannot push the hotfix back this way. Option A is the closest wrong answer: selective deployment exists, but only going forward. Deployment rules change settings such as connections between stages, and workspace roles control who can deploy; neither lifts the empty-target requirement. The sensible route is to make the fix in Development and deploy forward.

Question 2 (Implement and manage an analytics solution)

A company wants two kinds of people to be able to add colleagues to a Fabric workspace, with the same or lower permissions than their own. Which two workspace roles can do this? Choose two.

  • A. Viewer
  • B. Admin
  • C. Contributor
  • D. Member
  • E. Capacity contributor
Show the answer and explanation

Correct answers: B and D. Admin and Member can add people to a workspace with equal or lower permissions. Contributor, the closest wrong option, can create and edit items but cannot manage who has access. Viewer is read-only. Capacity contributor is not a workspace role at all: it controls who may assign workspaces to a capacity, not who can join a workspace. When a question asks about adding people, think about which roles manage access rather than which roles edit content.

Question 3 (Ingest and transform data)

A data engineer creates a OneLake shortcut in a lakehouse that points to files in an Amazon S3 bucket. The account used by the connection has full write permissions on the bucket. A notebook then tries to write result files through the shortcut. What happens?

  • A. The files are written to OneLake and copied to S3 later
  • B. The write succeeds, because the account can write to the bucket
  • C. The write succeeds only if the bucket is in the same region as the capacity
  • D. The write fails, because Amazon S3 shortcuts are read-only
Show the answer and explanation

Correct answer: D. Shortcuts to Amazon S3 are read-only in OneLake, no matter what permissions the connection account holds in S3. To save results, the notebook should write to the lakehouse itself or to a shortcut type that supports writes. Option B is the trap: it assumes the source permission decides the outcome, but the shortcut type sets the limit first. Caching on a shortcut only speeds up reads and does not queue writes back to the source, so option A is wrong, and region does not change the read-only behavior.

Question 4 (Ingest and transform data)

A streaming query must report the total number of readings every 15 minutes, using the previous 15 minutes of data, so that each reading is counted in exactly one result. Which window type fits?

  • A. Hopping window
  • B. Session window
  • C. Tumbling window
  • D. Sliding window
Show the answer and explanation

Correct answer: C. A tumbling window cuts the stream into fixed, back-to-back segments that never overlap, so each reading falls into exactly one 15-minute result. A hopping window is the closest wrong option. It advances by a hop that can be shorter than the window size, so events can fall into several windows. If the hop equals the window size, a hopping window behaves like a tumbling one, but the question gives no reason to choose it. Session windows depend on gaps in activity, and sliding windows produce output as events enter or leave.

Question 5 (Monitor and optimize an analytics solution)

An analyst connects to the SQL analytics endpoint of a lakehouse and tries four T-SQL statements. Which one can succeed?

  • A. INSERT INTO dbo.Orders (Id, Qty) VALUES (1, 5)
  • B. CREATE TABLE dbo.OrdersStage (Id INT, Qty INT)
  • C. UPDATE dbo.Orders SET Qty = 0 WHERE Id = 1
  • D. CREATE VIEW dbo.vOrders AS SELECT * FROM dbo.Orders
Show the answer and explanation

Correct answer: D. The SQL analytics endpoint of a lakehouse is read-only for table data, but you can still create objects such as views, functions, and stored procedures on top of the tables. Inserting, updating, or creating tables belongs to a Fabric data warehouse; for lakehouse tables you change data with Spark or another writer. The closest wrong option is B, because creating a table sounds like a schema change that a SQL user might expect to do. In a lakehouse, tables are created through the lakehouse itself instead.

More guides for DP-700

Next step

If you missed questions, go back to the domains explained page for that domain. For more practice at exam length, try our DP-700 practice tests.

Try the DP-700 practice exams