DP-900
Microsoft Azure Data Fundamentals
DP-900 Exam Guide: Azure Data Fundamentals Preparation and Study Roadmap
DP-900 validates foundational knowledge of data concepts, Azure data services, relational and non-relational storage, and analytics workloads. It is designed for beginners who are starting to work with data in the cloud rather than for candidates who already administer databases or build advanced pipelines. This guide helps you decide whether the certification matches your starting point, identify the skills that deserve the most study time, choose suitable Microsoft Learn resources, and schedule the assessment only after you can explain service choices in practical scenarios.
What DP-900 validates
DP-900 tests whether you can recognize core data concepts and match common Azure data services to relational, non-relational, and analytical requirements. Microsoft classifies the certification as Beginner level, with Azure as the product, Data Engineer as the role, and Databases as the subject. It is associated with Microsoft Certified: Azure Data Fundamentals. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
The certification is aimed at a candidate beginning to work with data in the cloud. Microsoft expects familiarity with relational and non-relational data concepts and with workloads such as transactional and analytical processing. You therefore need breadth more than implementation depth: know what a technology is for, what kind of data it handles, and why one workload may suit it better than another. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
This is a useful entry point if you work in application development, analysis, database support, cloud adoption, teaching, or an adjacent technical role and need a shared vocabulary for Azure data. It can also support later preparation for Azure Database Administrator Associate or Azure Data Engineer Associate certifications, but Microsoft states that DP-900 is not a prerequisite for either. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
Who should take it, and who should wait
Take DP-900 when you need a structured introduction to cloud data and can study the concepts behind several Azure services without needing to configure production systems. Wait or choose a more advanced path when your immediate goal is detailed database administration, complex data engineering, or specialist performance tuning. DP-900 establishes foundations; it does not replace role-specific experience. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
There is no prerequisite listed for the Microsoft Learn core data concepts path or the DP-900 academic preparation path. The certification page does, however, say that candidates should be familiar with the self-paced or instructor-led DP-900 learning material. Treat that as a readiness expectation rather than a formal prerequisite: you do not need an earlier certification, but you should not approach the exam without learning the vocabulary and service categories. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
A practical decision test is whether you can already distinguish a relational table from a document or key-value model, and a transactional workload from an analytical workload. If those distinctions are unfamiliar, begin with the fundamentals path. If they are familiar but Azure services are not, move through the service-focused material and use the practice assessment to locate gaps.
Which skills are measured
The current DP-900 study guide groups the assessment into four domains: Describe core data concepts (25–30%), Identify considerations for relational data on Azure (20–25%), Describe considerations for working with non-relational data on Azure (15–20%), and Describe an analytics workload on Azure (25–30%). Use these labels in your notes so every review question is tied to an official exam domain. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Describe core data concepts (25–30%) is not merely a glossary exercise. The study guide includes ways to represent data, structured, semi-structured, and unstructured data, data storage options, common file formats, common data stores, and Azure datastore choices. Prepare to classify an example before naming a service: first identify the shape and use of the data, then consider where it belongs. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Identify considerations for relational data on Azure (20–25%) requires a clear model of relational databases and the Azure services that support them. Review tables, columns, relationships, structured querying, and the distinction between a database engine and the cloud service hosting or managing it. The official course describes relational database concepts and relational database services in Azure as central topics. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
Describe considerations for working with non-relational data on Azure (15–20%) covers data that does not depend on a relational schema. Study the purpose of Azure Storage for non-relational information and the fundamentals of Azure Cosmos DB. The course also identifies non-relational data as a separate foundation from relational databases, so do not assume that every storage decision begins with tables and relationships. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
Describe an analytics workload on Azure (25–30%) concerns how data is used for analysis rather than only how it is stored. The course lists large-scale data warehousing, real-time analytics, and data visualization. Connect those ideas to the broader distinction between transactional and analytical workloads, then map the data movement, analysis, and presentation stages to appropriate Azure services. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
Microsoft describes the bullets beneath the measured skills as illustrations of assessment coverage and notes that related topics may also appear. Most questions cover generally available features, although preview features may be included when they are commonly used. For that reason, study the domain descriptions as boundaries and use the linked resources for current service terminology rather than relying on an old memorized list. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
How to use the domain weights
The percentages should determine study priority, not dictate a narrow cram strategy. Start with the two domains that each represent 25–30%—Describe core data concepts and Describe an analytics workload on Azure—while reserving enough time for relational and non-relational Azure considerations. Every percentage should remain attached to its named domain because the domains measure different kinds of judgment. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
A sensible allocation is to learn core concepts first, then relational and non-relational service families, and finish by tracing an analytics workload from data sources through storage and processing to visualization. This order is a practical recommendation, not a Microsoft requirement. It works because service comparisons are easier once you understand the data type and workload they serve.
Do not turn the weights into a prediction of how many questions you will see from each domain. Microsoft presents them as skill-measurement ranges, and the study guide says the listed bullets illustrate coverage rather than exhaustively defining every possible related topic. Use the ranges to protect study time for each domain, then use assessment feedback to adjust your final revision.
The Azure data concepts to learn first
Begin by classifying data and workload before studying product names. You should be able to explain structured, semi-structured, and unstructured data; distinguish relational from non-relational stores; recognize common file formats; and separate transactional processing from analytical processing. These distinctions give you the reasoning framework needed for later Azure service questions. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
For structured data, focus on predictable fields, tables, columns, and relationships. For semi-structured data, focus on records whose fields can be represented with tags, keys, or a flexible schema. For unstructured data, focus on content that does not naturally fit a fixed tabular model. These descriptions are study tools: the exam expects you to identify characteristics and storage implications, not recite a single definition without context.
Transactional workloads support operational activity such as recording or updating individual business events. Analytical workloads examine data to identify trends, compare results, or support decisions. Write a two-column comparison in your notes covering purpose, access pattern, and the type of result expected. Then add an example of a system that records activity and a separate process that analyzes accumulated activity.
Review data roles and the stages through which data moves. A data analyst, database administrator, data engineer, developer, and solution architect may interact with the same solution from different angles. The Microsoft Learn core concepts path includes data roles and services, making it useful for connecting responsibilities with technology choices. (https://learn.microsoft.com/en-us/training/paths/azure-data-fundamentals-explore-core-data-concepts/)
A common mistake is to memorize service names before understanding the problem they solve. Correct that by creating flashcards with the requirement on the front and the data category or workload on the back. For example, ask yourself whether a scenario is operational or analytical before asking which Azure service appears in the answer choices.
How to prepare for relational data on Azure
Learn relational data as both a modeling approach and a family of Azure options. You should understand tables, columns, relationships, and query-oriented access, then recognize that Azure offers several ways to run or manage relational database workloads. The objective is to identify the appropriate consideration, not to become a database administrator before sitting DP-900. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
Build a service comparison sheet from official Microsoft Learn material. For each relational option you study, record the type of workload it supports, the level of management it implies, and the situation in which an organization might choose it. Keep the entries conceptual and current. Avoid copying an undocumented feature matrix from a third-party question bank because service capabilities and product descriptions can change.
Use small scenarios to test understanding. Ask what happens when an application needs consistent records across related entities, when an existing SQL-based workload is moved to Azure, or when an organization wants a managed relational service instead of managing infrastructure itself. Explain the choice in one or two sentences; if your explanation only repeats the product name, the concept is not yet secure.
The official DP-900 course covers relational data in Azure as part of its beginner-level curriculum and can be taken through instructor-led training or self-paced study. Microsoft lists the course duration as 1 day, but that is the course listing rather than a promise that one day is enough for every candidate’s exam preparation. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
Do not spend most of your time writing SQL syntax unless it helps you understand the relational model. DP-900 is a fundamentals assessment. Prioritize the purpose of relational services, the concepts they implement, and the distinction between transactional and analytical use over lengthy query drills.
How to prepare for non-relational data and Cosmos DB
Non-relational preparation should focus on choosing a suitable data model and understanding Azure storage categories. Review key-value, document, and column-family models, along with the reasons an application may avoid a fixed relational architecture. Then study Azure Storage and Azure Cosmos DB as distinct parts of the non-relational landscape rather than treating NoSQL as one interchangeable service. (https://techcommunity.microsoft.com/blog/educatordeveloperblog/dp-900-microsoft-azure-data-fundamentals-study-guide/3791691)
For each model, write the shape of an example record and the way an application would locate it. A key-value model emphasizes a key and its associated value. A document model emphasizes a self-contained document. A column-family model organizes related values for access patterns that differ from traditional tables. The point is to recognize the model in a scenario and explain why its flexibility or access pattern matters.
Microsoft describes Azure Cosmos DB as a fast NoSQL database with SLA-backed speed and availability, automatic and instant scalability, and open-source APIs for MongoDB and Cassandra. Use those documented characteristics to understand where Cosmos DB fits, but do not convert them into a guarantee for an unspecified application or workload. (https://techcommunity.microsoft.com/blog/educatordeveloperblog/dp-900-microsoft-azure-data-fundamentals-study-guide/3791691)
A useful exercise is to compare a relational customer record with a document-oriented customer profile. Identify which information is fixed, which may vary, and whether the application needs relationships and joins or a flexible document representation. Then ask whether the scenario emphasizes globally scalable application access, simple object storage, or a database query model. This prevents the common error of labeling every non-tabular requirement “Cosmos DB.”
Hands-on practice is optional but useful. Microsoft’s learning paths offer Azure account options that include pay-as-you-go or an Azure free option for up to 30 days; check the current terms before creating resources, and remove anything you no longer need. (https://learn.microsoft.com/en-us/training/paths/azure-data-fundamentals-explore-core-data-concepts/)
How to study analytics workloads
Treat analytics as a chain of decisions: collect or access data, store it appropriately, process it for the required form of analysis, and present the result. DP-900 preparation should connect data warehousing, real-time analytics, and data visualization to that chain. If you study each service in isolation, you may recognize names but still miss the workload the question describes. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
Contrast large-scale analytical storage with operational databases. A warehouse-oriented workload is designed around analysis across accumulated information, whereas a transactional system records current business activity. Real-time analytics introduces a need to analyze incoming or rapidly changing data with limited delay. Visualization then communicates findings to people who need to interpret the results.
The official DP-900 material identifies Azure Data Factory and Azure Stream Analytics among the Azure data services covered in its study guidance, and the course includes analytics topics such as data warehousing, real-time analytics, and data visualization. Build a simple flow diagram that places each concept in relation to the others; do not memorize a service as the answer to every data movement or streaming scenario. (https://techcommunity.microsoft.com/blog/educatordeveloperblog/dp-900-microsoft-azure-data-fundamentals-study-guide/3791691)
Practice explaining why a scenario is batch or real-time, operational or analytical, and structured or non-relational. Those pairs are more useful than isolated definitions because exam questions often describe a business requirement before asking you to identify a workload or service category. If you cannot explain the requirement in plain language, return to the core concepts module before adding more product notes.
A frequent pitfall is studying visualization as a separate reporting tool and ignoring the data preparation that makes a report useful. Your notes should show what information is being analyzed, where it originates, how it is processed, and how a consumer sees the result. That sequence also helps distinguish analytics services from the database that stores operational records.
Which official resources to use
Use the Microsoft study guide as the control document, Microsoft Learn paths for structured learning, and the practice assessment and sandbox for readiness and interface familiarity. This combination covers scope, instruction, feedback, and exam-environment practice without depending on leaked questions or unsupported claims. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Start with Microsoft’s “Introduction to Microsoft Azure Data core data concepts” learning path. It has no prerequisites and includes modules on exploring core data concepts and exploring data roles and services. Complete it before attempting detailed service comparisons, because it supplies the vocabulary used by the later relational, non-relational, and analytics material. (https://learn.microsoft.com/en-us/training/paths/azure-data-fundamentals-explore-core-data-concepts/)
Use the official DP-900 course, “Introduction to Microsoft Azure Data,” as the organized syllabus when you prefer instructor-led learning or want to follow a self-paced course structure. It covers relational and non-relational data, big data, analytics, Azure Storage, Azure Cosmos DB, data warehousing, real-time analytics, and data visualization. (https://learn.microsoft.com/en-us/training/courses/dp-900t00)
The academic preparation path is primarily for educators preparing to teach DP-900 in academic programs. It contains modules about Microsoft Learn for Educators, course planning, delivery, labs, and student certification. It is relevant to instructors, but most individual candidates should prioritize the certification page, study guide, and candidate learning paths instead. (https://learn.microsoft.com/en-us/training/paths/prepare-teach-dp-900-fundamentals-academic-programs/)
Use the Microsoft practice assessment after one complete pass through the material, not as your only source of learning. Microsoft says practice assessments provide an overview of likely question style, wording, and difficulty and can help identify knowledge gaps. Record the concept behind every missed answer, then revisit the relevant domain rather than simply repeating the same test. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
Launch the exam sandbox before the assessment. Microsoft describes it as an experience demo that lets you interact with different question types in the same user interface used during the exam. This is a practical recommendation for reducing interface uncertainty; it does not reveal live exam content or predict the questions you will receive. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
A practical DP-900 study roadmap
A four-stage roadmap is more reliable than reading every Azure data page in random order: establish concepts, map services, connect workloads, and verify readiness. The timetable should reflect your existing experience and available study time. The sequence below is a practical preparation plan, not an official Microsoft schedule. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Stage 1: establish the vocabulary
Begin with the core data concepts learning path and create a one-page glossary. Include structured, semi-structured, and unstructured data; relational and non-relational stores; transactional and analytical workloads; common file formats; and the roles involved in a data solution. After each topic, write one scenario in your own words. This exposes confusion earlier than passive reading. (https://learn.microsoft.com/en-us/training/paths/azure-data-fundamentals-explore-core-data-concepts/)
Stage 2: map Azure services to requirements
Study relational and non-relational Azure material with a comparison table. Put the requirement in the first column, the data model in the second, the workload in the third, and candidate services in the fourth. Include a short reason for rejecting at least one alternative. This forces you to make the decision the exam is testing instead of associating a product name with a memorized keyword.
Stage 3: connect the analytics flow
Draw several end-to-end diagrams that begin with data generation and finish with analysis or visualization. Mark where operational storage, non-relational storage, data movement, streaming, warehousing, and reporting fit. Explain each diagram aloud or in writing. If the same service appears in every diagram regardless of the requirement, revisit the workload distinction and service purpose.
Stage 4: test and repair gaps
Take the official practice assessment and classify each missed or guessed item by domain. Re-study the concept, close the material, and answer a new scenario without looking at the explanation. Repeat the cycle until your reasoning is consistent. Then use the sandbox to familiarize yourself with the interaction model and review the study guide for updates before scheduling. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
How to know you are ready
Readiness means you can explain choices, not that you have memorized a service catalogue. Before scheduling, you should be able to classify data, distinguish workloads, describe the purpose of the major Azure data categories in your study material, and correct your own reasoning when a scenario changes. Use the official practice assessment as evidence of gaps, not as a pass guarantee. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
Run a closed-book review using four prompts, one for each named domain. For core data concepts, classify a data representation and workload. For relational data on Azure, explain why a relational service fits a requirement. For non-relational data on Azure, identify a suitable model and storage category. For an analytics workload on Azure, describe the path from source data to an analytical result.
You are not ready if your notes contain only definitions, if you confuse storage with processing, or if you choose Cosmos DB whenever a question mentions scale. You also need another review if you cannot distinguish an operational database from a warehouse-oriented workload or if you rely on a practice question’s wording rather than the underlying requirement.
Use the study guide’s update information before booking. Microsoft says the English version is updated first and that localized versions may be updated approximately eight weeks later, although the schedule is not guaranteed. Most questions cover generally available features, with possible coverage of commonly used preview features. Check the live study guide and exam page for the version relevant to your chosen language. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
What the delivery details mean for scheduling
Microsoft states that DP-900 is a proctored assessment and provides 45 minutes to complete it. It may include interactive components. Use the current certification page for the live exam experience and scheduling instructions, because delivery policies and availability can change. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
The listed exam languages are English, Japanese, Chinese (Simplified), Korean, French, German, Spanish, Portuguese (Brazil), Russian, Indonesian (Indonesia), Arabic (Saudi Arabia), Chinese (Traditional), and Italian. If the exam is not available in your preferred language, Microsoft says you can request an additional 30 minutes to complete it. Confirm language and accommodation arrangements through the official process before scheduling. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
You can schedule through Pearson VUE, or through Certiport if you are a student or educator. Microsoft strongly recommends registering with a personal Microsoft account because records tied to an organizational work or school account may be lost and unrecoverable if you leave that organization. Connect your certification profile to Microsoft Learn so you can schedule and renew exams and share or print certificates. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
The exam price depends on the country or region in which it is proctored. Do not rely on a universal price shown on an unofficial page; check the scheduling flow for your location. If you need accommodations, request them through Microsoft’s certification resources before the appointment rather than assuming the standard process covers your circumstances. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
On the day, allocate time so that difficult questions do not consume the entire assessment. Read the workload and data-model clues first, eliminate options that belong to a different category, and return to uncertain items if the interface permits. These are practical test-taking recommendations, not observations about undisclosed exam behavior.
Common preparation mistakes to avoid
The most damaging mistakes are studying outdated objectives, confusing related services, and treating practice questions as a substitute for understanding. A disciplined review of the current study guide, followed by scenario-based explanations, is more defensible than collecting increasingly large sets of unverified questions. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Mistake one is using an older blueprint without checking its domain labels. The current study guide lists Describe core data concepts (25–30%), Identify considerations for relational data on Azure (20–25%), Describe considerations for working with non-relational data on Azure (15–20%), and Describe an analytics workload on Azure (25–30%). Older material may use different names or ranges, so anchor revision to the current official guide. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Mistake two is memorizing product features without learning the workload. A question that mentions scale, flexibility, streaming, or reporting still requires you to identify the underlying requirement. Build explanations that begin with data shape and access pattern, then name the service category. This approach also makes it easier to adapt when Microsoft changes a product description.
Mistake three is treating every practice result as a score prediction. Microsoft presents practice assessments as tools for understanding question style, wording, difficulty, and knowledge gaps. Review why an answer is correct and why alternatives are wrong. Never rely on exam dumps, leaked questions, or memorization as a guarantee of passing. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
Mistake four is ignoring the exam interface until the appointment. The official sandbox lets you experience the look and feel of the exam and interact with different question types. Use it after learning the content so the sandbox supports confidence with navigation rather than becoming another source of unfocused study. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
Mistake five is booking with an account that you may lose access to. Use a personal Microsoft account as Microsoft recommends, and confirm that your certification profile is connected before completing registration. This administrative step is easy to overlook and difficult to repair after an organizational account becomes unavailable. (https://learn.microsoft.com/en-us/credentials/certifications/azure-data-fundamentals/)
What to do after a first attempt
Use the score report to identify concepts that need repair, then return to the relevant official domain rather than restarting the entire syllabus automatically. Microsoft requires a score of 700 or greater to pass DP-900. If you do not pass, Microsoft says you can retake the exam after 24 hours following the first attempt; later retake intervals vary, so check the current retake policy before making another appointment. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
For a weak core concepts result, rebuild the data-model and workload glossary. For a weak relational or non-relational result, make service comparisons from the official learning material and practice scenario explanations. For a weak analytics result, redraw the end-to-end flow and distinguish operational, batch, real-time, warehouse, and visualization requirements. This targeted repair is more efficient than rereading material you already understand.
Do not interpret a failed attempt as evidence that more memorization alone will solve the problem. Identify whether the issue was a missing concept, a confused service boundary, an outdated resource, or poor time allocation. Take another official practice assessment only after repairing the cause, and use the sandbox again if interface familiarity contributed to the difficulty.
Your next actions
Start with the current DP-900 study guide, note the four domains and their ranges, and complete the core concepts learning path. Next, work through relational, non-relational, and analytics material in a comparison-oriented notebook. Finish with the official practice assessment and exam sandbox, then verify language, timing, account, accommodation, and scheduling details on Microsoft’s live certification page. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Before booking, write a short explanation for each of these decisions: when relational data is appropriate, when a non-relational model is appropriate, how a transactional workload differs from an analytical one, and how an Azure analytics solution moves from data to insight. If your explanations are specific and survive changes in the scenario, schedule. If they are still product-name lists, return to the relevant learning path.
Keep the study guide in your final review folder. Microsoft updates the English exam version first and provides the current measured-skills information there, so it is the best place to check for changes before the appointment. The certification page remains the source for current delivery, language, scheduling, and retake information. (https://learn.microsoft.com/en-us/credentials/certifications/resources/study-guides/dp-900)
Conclusion
DP-900 is best approached as a reasoning exam for cloud data fundamentals. Learn the data models and workloads first, connect them to Azure service categories, and use the official assessment tools to expose gaps. Keep each study decision tied to one of the four named domains, verify time-sensitive details on Microsoft Learn, and schedule only when you can explain why a service fits a requirement rather than merely recognize its name.
Related exams
- AZ-104 exam — Microsoft Azure Administrator
- AZ-120 exam — Planning and Administering Microsoft Azure for SAP Workloads
- AZ-140 exam — Configuring and Operating Windows Virtual Desktop on Microsoft Azure
- AZ-305 exam — Designing Microsoft Azure Infrastructure Solutions
- AZ-400 exam — Microsoft Azure DevOps Solutions
- AZ-500 exam — Microsoft Azure Security Technologies