The case study can open the exam.
Passers who expected the case study last were surprised to get it as the first group of questions, with a lot of detail. One suggests sketching the requirements on the digital whiteboard.

Validates knowledge of preparing data and building semantic models in Microsoft Fabric, including governance, SQL, KQL and DAX.
DP-600 tests whether you can prepare, secure and model data for analysis in Fabric through scenarios and a case study. Build lakehouse, warehouse and semantic model work yourself in a Fabric trial, practise SQL, KQL and DAX queries, and give security and Direct Lake real study time.
Example scenario: Sales managers must see only their region's rows in a shared semantic model, and finance must not see a margin column at all. What would you configure?
Example scenario: A lakehouse table has duplicate customer rows with different update times. How would you build a clean dimension table from it in a warehouse?
Example scenario: A Direct Lake model falls back to DirectQuery and reports slow down. What would you check, and how could you stop the fallback?
DP-600 study guide · Skills measured as of October 19, 2026
Choose the best answer from the options, or the stated number of answers when a question asks for more than one.
Each sales manager must see only their own region's rows in a shared semantic model. What should you configure?
A. Row-level security filters rows by role. Object-level security hides tables or columns, not rows.
Move items from a list into the correct places, such as matching items to targets or putting steps in order.
Drag each task to the language best suited to it.
Write a measure → DAX; query the SQL analytics endpoint → T-SQL; explore events in an eventhouse → KQL.
Choose the correct option for each part of an exhibit or statement list, often Yes or No for each statement.
For each statement, select Yes or No.
Yes, No, Yes. The lakehouse SQL analytics endpoint is read-only.
Read one scenario with its background, requirements and exhibits, then answer several questions about it.
Case study: A retailer keeps order history in a Fabric lakehouse and wants reports that read it without copying the data into the model. Question 1 of 3: Which storage mode should the semantic model use?
A. Direct Lake reads Delta tables in OneLake directly, so there is no import copy to refresh.
Sign in to Microsoft Learn with a personal Microsoft account, complete your certification profile, then schedule DP-600 with Pearson VUE.
Take the free practice assessment to see where you start, then work through the official learning paths and build each item in a Fabric trial.
Most candidates spend 40–80 hours. Plan for the low end if you already build Fabric semantic models; add time if Fabric is new to you or you are also new to SQL.
The free learning paths cover the skills measured. A video course helps if you prefer watching demos. Use a Fabric trial for hands-on practice.
Do not skip security. Passers say workspace and item security, T-SQL and semantic model design carry more weight than basic KQL. Practise each in your trial tenant.
Start with Microsoft's free practice assessment. MeasureUp's official test adds timed sets. Several candidates scored high on practice and still failed, so review every explanation and rebuild weak areas.
Choose the checklist for your test-centre or online appointment.
Passers who expected the case study last were surprised to get it as the first group of questions, with a lot of detail. One suggests sketching the requirements on the digital whiteboard.
One candidate failed with 672 despite 80% or more on MeasureUp, then passed with 850 after focused review. Another scored 90 to 100% on practice and still failed.
A passer new to Fabric scored 838 after two to three weeks of daily study and said not to skip the Microsoft Learn exercises. Others say KQL questions are basic while T-SQL, security and semantic models carry more weight.