If you have been building Power BI reports at work for a year and keep telling yourself you will book the PL-300 exam "once things quieten down", this is the post that gets it in the diary. The PL-300 exam is passable in eight weeks of structured evening study, but only if you study the objectives Microsoft actually publishes rather than the ones most blog posts still repeat.
That distinction matters more than usual right now. Microsoft refreshed the PL-300 skills measured document on 20 April 2026, and a striking number of the guides currently ranking on Google are quoting domain weights that no longer exist. If you build a revision timetable around those numbers, you will over-invest in data modelling and walk into the exam light on data preparation and governance.
This plan uses the current objectives, sets out what the exam looks like on the day, and gives you a week-by-week schedule you can actually follow alongside a full-time job.
What the PL-300 Exam Covers in 2026
PL-300 earns you the Microsoft Certified: Power BI Data Analyst Associate credential. It is a single exam, with no prerequisite certification, and it sits at associate level.
As of 20 April 2026, Microsoft measures four skill areas:
| Skill area | Weighting | What it really tests |
|---|---|---|
| Prepare the data | 25 to 30% | Power Query, connecting to sources, choosing between Import, DirectQuery and DirectLake, profiling and cleaning |
| Model the data | 25 to 30% | Relationships, cardinality, star schema design, DAX measures, time intelligence, performance tuning |
| Visualize and analyze the data | 25 to 30% | Choosing and formatting visuals, bookmarks, drillthrough, accessibility, AI visuals, Copilot |
| Manage and secure Power BI | 15 to 20% | Workspaces, apps, gateways, scheduled refresh, row-level security, sensitivity labels |
Exam Tip: The three largest domains are weighted almost identically at 25 to 30% each. There is no single dominant domain to specialise in, so any plan that spends half its time on DAX is misallocating your revision. Split your effort roughly evenly across the first three areas, then give governance its own dedicated block.
This is the detail that trips people up. A lot of older PL-300 content describes "Model the data" as a 30 to 35% giant and "Prepare the data" as a 15 to 20% afterthought. That was true of a previous version of the exam. It is not true of the exam you will sit in 2026, and data preparation now carries the same weight as modelling.
The Copilot and Fabric additions nobody warns you about
The April 2026 objectives quietly folded generative AI into the exam. Under "Visualize and analyze the data", Microsoft now explicitly lists creating a narrative visual with Copilot, using Copilot to create a new report page, using Copilot to suggest content for a report page, and using Copilot to summarise the underlying semantic model.
Under "Prepare the data", the objectives now ask you to choose between DirectLake, DirectQuery and Import. DirectLake is a Microsoft Fabric storage mode, so PL-300 has quietly absorbed a slice of Fabric knowledge that did not appear in earlier versions of the exam.
If your preparation material predates spring 2026, it almost certainly does not cover either. Budget explicit study time for both. They are cheap marks because they are new, and new objectives are consistently the ones candidates skip.
PL-300 Exam Format at a Glance
| Detail | Value |
|---|---|
| Exam code | PL-300 |
| Certification | Microsoft Certified: Power BI Data Analyst Associate |
| Questions | Typically 40 to 60 |
| Duration | Around 100 minutes |
| Passing score | 700 out of 1000 |
| Cost | 165 US dollars (varies by region) |
| Delivery | Pearson VUE test centre or OnVUE online proctored |
| Prerequisites | None |
| Renewal | Annually, free, via an unproctored Microsoft Learn assessment |
A score of 700 or greater is required to pass the PL-300 exam. That is a scaled score, not a percentage, so do not translate it into "I need 70% of questions right". Microsoft scales to account for difficulty variation between question sets, and the number of questions you can afford to miss shifts accordingly.
The annual renewal is genuinely painless. Once you hold the certification, Microsoft emails you within six months of expiry and you renew by passing a free, open-book, unproctored assessment on Microsoft Learn. There is no re-sit fee and no need to return to a test centre.
If you are taking the exam from home, read our guide to OnVUE online proctored exam rules before booking. The check-in process and room scan catch out more candidates than the exam content does.
Is the PL-300 Exam Hard? Where Candidates Actually Lose Marks
PL-300 has a reputation as one of the friendlier Microsoft associate exams, and relative to something like AZ-305 that is fair. It is not, however, a fundamentals exam. It is scenario-based, and the questions describe a business situation and ask what you would do rather than asking what a feature is called.
Ask around and everyone will tell you DAX is the hardest part. They are half right. DAX is the most intimidating part, but it is not where most marks are lost. Three areas do more damage:
- Power Query transformations. Merge versus append, reference versus duplicate queries, pivot and unpivot, and converting semi-structured data into a table. These questions are fiddly and they are worth 25 to 30% of the exam.
- Row-level security. RLS roles, group membership and item-level access come up reliably, and candidates who only build reports for themselves have often never configured any of it.
- Storage mode decisions. Knowing when Import beats DirectQuery, and now when DirectLake beats both, is a recurring scenario theme.
Exam Tip: If you have only ever used Power BI Desktop and have never published to a workspace, configured a gateway, or set up a scheduled refresh, the "Manage and secure Power BI" domain will hurt. That is 15 to 20% of the exam sitting in a part of the product you may never have opened. Get a free Fabric trial and actually publish something.
The honest test of readiness is this: can you look at a scenario, decide on a storage mode, sketch a star schema, and explain which DAX pattern solves the requirement, without touching a keyboard? If yes, you are close.
The 8-Week PL-300 Study Plan
This plan assumes roughly eight to ten hours a week, split across weekday evenings and one longer weekend session. If you already work in Power BI daily, you can compress it to six weeks by merging weeks two and three. If you are new to the tool, stretch it to twelve.
Week 1: Foundations and honest baseline
Install Power BI Desktop, sign up for a free Microsoft Fabric trial, and take a full practice assessment before you study anything. A cold-start score of 40 to 50% is normal and it tells you which domain to prioritise.
Work through Microsoft Learn's Power BI data analyst learning path introduction, and download a sample dataset you will use for the rest of the plan. Consistency matters here. Using one dataset throughout means you spend your time learning features, not re-learning data.
Week 2: Prepare the data, part one
Focus entirely on connecting and profiling. Practise connecting to at least four different source types, changing data source settings and credentials, and setting privacy levels. Spend real time in the column quality, column distribution and column profile panes.
Then learn the storage modes properly. Write yourself a one-page decision table covering Import, DirectQuery and DirectLake, with the trade-offs for data volume, refresh latency and query performance. This is the single highest-value page of notes in the whole plan.
Week 3: Prepare the data, part two
This is Power Query week. Work through data types, creating and transforming columns, grouping and aggregating, pivot and unpivot and transpose, and converting semi-structured data such as JSON into a table.
Then drill the four operations that generate the most exam confusion: merge, append, reference and duplicate. Build examples of each and write down in your own words what changes downstream when you use one rather than another.
Finish the week by building proper fact and dimension tables from a flat file. You will need them next week.
Week 4: Model the data, part one
Relationships and structure. Cardinality, cross-filter direction, role-playing dimensions, and a common date table. Build the date table with DAX rather than importing one, because you need to be able to reason about it.
Configure table and column properties, and learn when a calculated column is appropriate versus when a measure is the right answer. Exam scenarios test that judgement repeatedly.
Week 5: Model the data, part two (DAX)
Now the part you have been dreading. Take it in this order:
- Single aggregation measures
- CALCULATE, and filter context generally
- Time intelligence measures
- Basic statistical functions
- Semi-additive measures
- Calculation groups and visual calculations
Do not try to memorise function signatures. The exam tests whether you can pick the right pattern for a requirement, so practise reading a business sentence and naming the DAX approach it calls for.
Close the week with performance: use Performance Analyzer and DAX query view to find slow measures and visuals, and practise reducing granularity to speed up a model.
Week 6: Visualize and analyze the data
Visual selection and formatting, themes, conditional formatting, slicing and filtering. Then the usability features that carry disproportionate exam weight: bookmarks, custom tooltips, visual interactions, report navigation, sync slicers, drillthrough with filters and buttons, and the Selection pane.
Set aside a dedicated session for accessibility and mobile layouts. Both are explicitly in the objectives and both are commonly skipped.
Then spend a full session on Copilot. Create a narrative visual with Copilot, use it to generate a report page, and use it to summarise a semantic model. Finish with AI visuals, grouping, binning, clustering, reference lines, forecasting and anomaly detection.
Week 7: Manage and secure Power BI
Publish your model to a workspace and treat this week as an admin week rather than an analyst week.
Create and configure a workspace, assign workspace roles, build and update an app, and choose a distribution method. Configure subscriptions and data alerts, promote and certify content, and configure a scheduled refresh. Understand exactly when a gateway is required.
Then implement row-level security end to end: create roles, write the DAX filters, test with "View as", and configure group membership. Finish with item-level access, semantic model access and sensitivity labels.
Week 8: Consolidation and exam readiness
No new material. Take a full practice exam early in the week, then spend the remaining sessions on your weakest domain only.
Re-read the Microsoft Learn skills measured document line by line and rate yourself on every bullet. Anything you rate below confident becomes a targeted revision block. Take a final practice exam two days before the real thing, and rest the day before.
Exam Tip: Book the exam at the start of week one, not at the end of week eight. A booked date is the single most reliable predictor of whether people actually finish a study plan.
Exam Day Tactics That Move Your Score
- Read the last line of the scenario first. PL-300 questions often bury three paragraphs of business context above a single specific requirement. Find the requirement, then read the context for constraints.
- Watch for the qualifier. "Minimise administrative effort", "minimise refresh time" and "minimise cost" change which of several technically correct answers is right.
- Do not fight a case study. If a scenario is consuming time, mark it, move on, and come back. There is no partial credit for a long fight.
- Answer everything. There is no negative marking, so an educated guess is strictly better than a blank.
- Treat Copilot and DirectLake questions as gifts. They are new, most candidates skimmed them, and they are usually straightforward if you revised them.
How PL-300 Fits Your Wider Certification Path
PL-300 has become more valuable, not less, as Microsoft has thinned out the Power Platform certification range. Several sibling exams have been retired or replaced during 2026, and PL-300 is one of the survivors, which concentrates recruiter recognition onto it. Our roundup of every Microsoft certification retiring in 2026 has the full list and dates.
Where you go next depends on your direction of travel:
- Deeper into data engineering: DP-700 is the natural progression, and the DirectLake content in PL-300 is a genuine head start. See our DP-700 six-week study plan.
- Broader across cloud: AZ-900 gives you the Azure vocabulary that data roles increasingly assume. Our AZ-900 four-week plan is the fastest route.
- Vendor-neutral data credentials: CompTIA's data range is worth a look alongside a Microsoft badge. Browse the full CertCrush course catalogue to compare.
Frequently Asked Questions
Is the PL-300 exam difficult?
PL-300 is moderately difficult and sits comfortably below expert-level exams like AZ-305. The challenge is not memorisation but application: questions describe a business scenario and ask which Power BI approach solves it. Candidates with a year of hands-on Power BI experience typically pass after six to eight weeks of structured study, while candidates who have only watched tutorials tend to struggle with Power Query and row-level security.
What score do I need to pass PL-300?
You need a scaled score of 700 out of 1000 to pass the PL-300 exam. This is not the same as scoring 70%, because Microsoft scales results to account for difficulty differences between question sets. You will see your pass or fail result on screen immediately after finishing, along with a bar chart showing your relative performance in each skill area.
How much does the PL-300 certification cost?
The PL-300 exam costs 165 US dollars in most regions, though the price varies by country and local taxes may apply. Renewal is free: Microsoft associate certifications expire annually and you renew by passing an unproctored online assessment on Microsoft Learn at no cost. Retakes are charged at the full exam price.
Can I take the PL-300 exam at home?
Yes. PL-300 is available through OnVUE online proctoring as well as at Pearson VUE test centres. Taking it at home requires a webcam, a stable internet connection, a government-issued photo ID and a completely clear desk and room, and you will complete a room scan before starting. The rules are strictly enforced, so read them properly before booking.
What if I fail the PL-300 exam?
If you fail, you must wait 24 hours before your first retake. After a second failure, the waiting period rises to 14 days between attempts, and you are limited to five attempts on the same exam within any twelve-month period. Each retake is charged at full price, so use the score report's skill-area breakdown to target revision rather than re-sitting quickly.
Ready to Start Practising?
Reading objectives will get you familiar with the PL-300 exam. Answering scenario questions under time pressure is what gets you past 700.
CertCrush gives you exam-style practice questions with detailed explanations for every answer, so you learn why the right option is right rather than just memorising it. Track your performance by domain, find the gaps in your preparation while there is still time to fix them, and walk into the test centre knowing what a pass feels like.
Create your free CertCrush account and start practising today, or browse the full course catalogue to plan your next certification after PL-300.