1.2 Exam Logistics & Blueprint
Key Takeaways
- DP-600 costs about US$165 in the United States (regional pricing varies) and is delivered by Pearson VUE, online-proctored or at a test center.
- Scoring is scaled to 1000; you need 700 to pass, and Microsoft does not publish exam-level pass-rate percentages.
- Microsoft role-based exams typically contain 40-60 questions; DP-600 gives you about 100 minutes and may include a case-study section.
- Prepare data carries the largest weight at 45-50% — allocate roughly half of all study time there.
- Microsoft Learn is accessible during the exam, but the clock keeps running, so treat it as a quick lookup, not a study substitute.
Core Exam Facts
Quick Answer: DP-600 costs US$165 (varies by region), runs about 100 minutes, is scored on a scaled 0-1000 range with 700 to pass, and typically contains 40-60 questions. Microsoft does not publish a public pass rate.
| Detail | Value |
|---|---|
| Exam code | DP-600 |
| Credential | Microsoft Certified: Fabric Analytics Engineer Associate |
| Delivery | Pearson VUE — online proctored or test center |
| Cost | ~US$165 (regional pricing varies) |
| Time | ~100 minutes |
| Questions | Typically 40-60 (Microsoft role-based exam norm) |
| Scoring | Scaled 0-1000; 700 required to pass |
| Pass rate | Not published by Microsoft |
| Skills version | Outline dated April 20, 2026 |
| Renewal | 12 months; free online Microsoft Learn assessment |
| Retake | 24-hour wait after first fail; 14-day wait for attempts 2-5; max 5 attempts in 12 months |
The scaled score means a raw percentage does not map linearly to 700. Question difficulty is statistically weighted, so 700/1000 is not the same as answering 70% of items correctly — a missed cluster of hard items hurts more than the count suggests. Aim well above a borderline performance on practice sets. If a question contains a Preview feature, note that Microsoft states most questions cover general-availability features, so default to GA behavior when unsure.
Question Formats and Case Studies
DP-600 mixes several interaction styles. Expect standard multiple choice plus interactive items:
- Drag-and-drop / build-list — order steps in a deployment pipeline or transformation sequence.
- Hot area — click the correct region of a screenshot (for example, the right workspace setting).
- Dropdown / yes-no statements — evaluate each statement independently for correctness.
- Case study — a fixed business scenario with multiple linked questions and tabs for requirements, existing environment, and data.
In a case study you can review the scenario tabs before and while answering, so read the requirements carefully before committing. Some items are unscored pilot questions; you cannot tell which, so answer everything as if it counts. Microsoft Learn documentation is available during the associate exam, but the timer does not pause while you browse it. Use it only to confirm an exact name or limit, never to learn a concept mid-exam.
The Three Weighted Domains — Study Allocation
The skills-measured outline (as of April 20, 2026) is organized into three domains. Watch out: some third-party prep sites still list a four-domain version with "Explore and analyze data" as a separate area — that is stale. Trust the official outline below and let the weights drive your hours.
| Domain | Exam Weight | Study Share | Focus |
|---|---|---|---|
| Maintain a data analytics solution | 25-30% | ~25-30% | Workspace/item access, row/column/object/file-level security, sensitivity labels, endorsement, Git, .pbip, deployment pipelines, XMLA, impact analysis |
| Prepare data | 45-50% | ~45-50% | Connections, OneLake catalog/Real-Time hub, store selection, views/functions/procs, star schema, denormalize/aggregate/join, SQL/KQL/DAX/Visual Query queries |
| Implement and manage semantic models | 25-30% | ~25-30% | Storage modes, relationships/bridge tables, advanced DAX, calculation groups, field parameters, large model format, Direct Lake, composite models, incremental refresh |
Most candidates with Power BI or Fabric experience need roughly 60-90 study hours over 4-8 weeks. Because Prepare data is the largest single block of marks, neglecting lakehouse/warehouse/eventhouse selection and transformation skills is the most common way well-prepared modelers still fail. A practical rule of thumb: out of an 80-hour plan, spend about 38 hours on Prepare data, 21 on semantic models, and 21 on maintaining the solution — then re-balance toward your weakest practice-test domain.
Registration, Day-of-Logistics, and Retakes
You register through your Microsoft Learn profile, which hands off to Pearson VUE for scheduling. Choose online proctored (taken from home with a webcam, a clear desk, and a stable connection) or an in-person test center. For online delivery you must verify identity, show the room to the proctor, and keep your face visible; phones, notes, and second monitors are prohibited. If the exam is not offered in your language, you can request an extra 30 minutes.
The retake policy is strict and worth memorizing because rushing a re-sit wastes a fee:
| Situation | Rule |
|---|---|
| Fail attempt 1 | Wait 24 hours before rebooking |
| Fail attempts 2-5 | Wait 14 days between each attempt |
| Maximum attempts | 5 per 12-month period |
| Passed already | Cannot retake the same version to improve a passing score |
Pacing on Exam Day
With about 100 minutes for roughly 40-60 questions, you have a little under two minutes per item on average — but the case study can consume a large block at once, so budget for it. A reliable strategy: do the standalone questions first at a steady pace, flag anything that needs a second look (the interface lets you mark and return), then attack the case study with the time you have banked. Never leave a question blank; there is no penalty for wrong answers, so an educated guess always beats an empty response.
Reserve the final five minutes to clear every flag before the timer ends, since you cannot return to a section once it is submitted in segmented case-study delivery.
A candidate has 80 study hours available and is strong at Power BI semantic modeling but new to Fabric. Based on the DP-600 blueprint weights, where should the largest single block of those hours go?
On exam day a candidate sees a question about a Microsoft Fabric feature they recall is still in Preview and is unsure how to answer. What is the most reliable approach?