DP-600 vs DP-700: which exam should you take first?
DP-600 and DP-700 are the two Microsoft Fabric associate exams. Both cover lakehouses, warehouses and pipelines, but they test different jobs: DP-600 is for the analytics engineer who models data for reporting, and DP-700 is for the data engineer who loads, monitors and tunes it.
Side by side
| DP-600 | DP-700 | |
|---|---|---|
| Certification | Microsoft Certified: Fabric Analytics Engineer Associate | Microsoft Certified: Fabric Data Engineer Associate |
| Questions | about 40–60 | about 40–60 |
| Time limit | 100 minutes | 100 minutes |
| Passing score | 700 on a scale of 1–1,000 | 700 on a scale of 1–1,000 |
| Exam fee | $165 USD (varies by country) | $165 USD (varies by country) |
| Valid for | 12 months, renewable free online | 12 months, renewable free online |
| Domains |
|
|
Who each exam is for
DP-600 is for analytics engineers who turn data into reusable analytics assets in Microsoft Fabric (lakehouses, warehouses, notebooks, dataflows, pipelines, semantic models and reports), with experience in SQL, DAX and PySpark.
DP-700 is for data engineers who ingest, transform, secure, monitor and optimize analytics data in Microsoft Fabric, and who work with SQL, PySpark and KQL.
Choose DP-600 if
- You come from Power BI and already know DAX and semantic models.
- Your work ends in reports: star schemas, Direct Lake models, measures and row-level security.
- You want the exam where preparing data is about half the marks and semantic models are a full domain.
Choose DP-700 if
- You build and run pipelines: full and incremental loads, orchestration, streaming.
- You are comfortable with PySpark and want KQL and real-time data on the exam.
- You care about operations: monitoring, alerts, troubleshooting and performance tuning are a third of the marks.
What overlaps
Study these once and they count for both exams.
- Choosing between a lakehouse, a warehouse and an eventhouse
- Ingesting data with pipelines, dataflows, notebooks and shortcuts
- Workspace and item access, row- and column-level security, dynamic data masking
- Git integration and deployment pipelines
- Sensitivity labels and endorsement
Only on DP-600
- Semantic models: storage modes and Direct Lake, relationships, DAX
- Calculation groups, field parameters, large and composite models
- Incremental refresh and model optimization tools
- Querying with the visual query editor
Only on DP-700
- Streaming with eventstreams, Spark Structured Streaming and KQL windows
- Mirroring, and full versus incremental loading patterns in depth
- Monitoring, alerts and troubleshooting for every item type
- Optimizing lakehouse tables, warehouses, eventhouses and Spark
Which to take first
If you come from Power BI or PL-300, take DP-600 first: its semantic-model domain builds on what you know, and its data-preparation domain introduces the Fabric items DP-700 goes deeper on. If you are a data engineer who writes Spark and SQL every day and rarely touches DAX, go straight to DP-700.
Start studying, free
Skillmarked is independent and not affiliated with Microsoft or Anthropic. Exam details change; check the official exam pages before you book.