DAMA CDMP in 2026: One Exam, Three Score Targets
The DAMA Certified Data Management Professional path is different from most certification programs because the Fundamentals exam is not simply pass or fail in the usual sense. The same 100-question exam can support Associate, Practitioner, or Master-level progress depending on your score and your broader credential path. That scoring structure changes how you should study. A candidate aiming for Associate can survive more weak areas than a candidate trying to reach Practitioner or Master on the first sitting.
Use DAMA's official CDMP certification page and the current CDMP exams page as your baseline. The local exam metadata lists the Fundamentals exam as 100 multiple-choice questions in 90 minutes, delivered online through Honorlock, with a $311 USD exam fee and optional Pay If You Pass availability. It also lists 60% for Associate, 70% for Practitioner, and 80% for Master-level scoring on the Fundamentals exam.
CDMP Fundamentals Snapshot
| Item | 2026 detail from local metadata |
|---|---|
| Credential body | DAMA International |
| Exam | CDMP Fundamentals |
| Questions | 100 multiple-choice questions |
| Time limit | 90 minutes, or 110 minutes with ESL accommodation |
| Delivery | Online proctored exam through Honorlock, available 24/7 once enrolled |
| Fee | $311 USD per exam |
| Scoring levels | 60% Associate, 70% Practitioner, 80% Master |
| Validity | Each passed exam valid for 3 years |
| Recertification | 120 continuing-education contact hours per 3-year cycle |
| Primary reference | DAMA-DMBoK 2, commonly called the Big Yellow Book |
The major strategic decision is your target score. If you only need Associate, your goal is broad competence and careful time management. If you are building toward Practitioner, you need 70% on Fundamentals and additional specialty exams. If you are aiming for Master, you need a much higher score, specialty exams at the higher threshold, and the experience assessment. Study for the credential path you actually want, not for the minimum score that makes a practice site look easy.
The CDMP Knowledge Area Map
The local metadata lists 14 knowledge areas. The four 11% domains are the first priority because together they represent 44 questions in a 100-question exam.
| Knowledge area | Weight | What to master |
|---|---|---|
| Data Governance | 11% | Stewardship, operating models, policies, standards, governance council, decision rights |
| Data Modeling and Design | 11% | Conceptual, logical, physical models, ERD notation, normalization, dimensional modeling |
| Data Quality | 11% | DQ dimensions, rules, measurement, DMAIC, root-cause analysis, remediation |
| Metadata Management | 11% | Business, technical, operational metadata, lineage, repositories, architecture patterns |
| Master and Reference Data | 10% | Golden records, MDM styles, reference data, match-merge, identity resolution |
| Data Warehousing and BI | 10% | Inmon, Kimball, star and snowflake schemas, ETL/ELT, SCDs, OLAP |
| Architecture, integration, security, document/content, storage/ops | 6% each | Enterprise architecture, ETL, privacy, ECM, database operations |
| Big data, ethics, process and maturity | 2% each | Data lakes, bias, fairness, maturity models, environmental factors |
The table shows why candidates fail after reading only the governance chapter. Governance is important, but it is one of several important areas. CDMP asks whether you can place a data problem in the right management discipline and choose the role, process, deliverable, or architecture that fits.
CDMP Traps That Cost Real Points
The first trap is role confusion. A data steward is usually accountable for business meaning, definitions, valid values, and quality expectations. A data custodian or DBA operates systems. A data architect designs structures. A governance council approves policies and standards. Many CDMP questions are role-boundary questions wearing scenario clothing.
The second trap is mixing model levels. Conceptual models communicate business concepts and relationships. Logical models add attributes, entities, relationships, and normalization independent of a specific platform. Physical models implement storage choices. If a question asks for business alignment, do not jump to indexes and partitioning. If it asks for implementation performance, a high-level concept map is too early.
The third trap is treating metadata as a single bucket. Business metadata explains meaning, ownership, rules, and terms. Technical metadata explains structures, formats, mappings, and lineage. Operational metadata explains processing, jobs, logs, and usage. Repository architecture can be centralized, distributed, or hybrid. These distinctions are easy points when memorized and expensive when blurred.
The fourth trap is confusing Inmon and Kimball. Inmon emphasizes an enterprise data warehouse orientation and normalized integrated data. Kimball emphasizes dimensional modeling, conformed dimensions, and business-process marts. Distractors often attach Kimball terms to Inmon architecture or reverse the flow.
The fifth trap is underestimating the 2% domains. Big Data, Ethics, and Maturity are small individually, but together they can swing a borderline score. For Master-level targets, there are no throwaway domains.
Study Order for a Strong First Attempt
Start with the official pages and the DMBoK table of contents. Build a one-page crosswalk with all 14 knowledge areas, their weights, their main roles, and their common deliverables. CDMP is easier when you can identify which knowledge area a scenario belongs to before reading the answer choices.
Then study the four 11% areas: governance, modeling, quality, and metadata. These are not isolated chapters. Governance defines accountability, modeling defines structure, quality defines trust, and metadata defines meaning and lineage. A real enterprise data issue often touches all four, so practice explaining the boundary between them.
Next study the two 10% areas: Master and Reference Data Management plus Data Warehousing and Business Intelligence. For MDM, memorize registry, consolidation, coexistence, and transactional patterns, plus golden record and match-merge logic. For warehousing and BI, compare Inmon and Kimball, then drill dimensions, facts, slowly changing dimensions, OLAP, ETL, and ELT.
After that, cover the 6% group: architecture, document and content management, integration and interoperability, security, and storage and operations. These are broad but testable. Make small comparison tables: RBAC versus ABAC, ETL versus ELT versus CDC, records management versus ECM, relational versus NoSQL, backup versus recovery, and architecture principle versus implementation standard.
Finish with big data, data science, ethics, and maturity. These are small in weight but high in conceptual language. Know volume, velocity, variety, veracity, data lake risks, bias and fairness, transparency, the DMBoK environmental factors, CMMI-DMM, DCAM, and maturity assessment goals.
How to Use OpenExamPrep Practice
A practical routing plan is diagnostic mixed questions first, then core domains, then middle domains, then small domains, then timed finish. When reviewing misses, label them by type: role boundary, framework vocabulary, architecture pattern, lifecycle sequence, definition, or scenario interpretation. Rereading the whole DMBoK is not efficient if your misses are mostly role-boundary errors.
Exam-Day Pacing
You have 90 minutes for 100 questions, which is less than one minute per item. The exam rewards quick classification. On each question, identify the knowledge area, then the role or process, then the requested deliverable or decision. If you cannot classify the scenario in 20 seconds, flag it and move. A candidate who spends four minutes debating an obscure data ethics item may sacrifice two easier governance or modeling points later.
For score targets, build margin. If your practice scores hover around 62%, you are not comfortably Associate-ready. If you need Practitioner, practice in the upper 70s before testing. If you need Master-level performance, low 80s in untimed practice is not enough; you need timed consistency because fatigue and close wording matter.
Official-Source Verification
Before sitting, verify fee, delivery, Pay If You Pass availability, ESL timing, recertification rules, and credential-level rules directly through DAMA and cdmp.info. Use the DAMA-DMBoK 2 as the content anchor. Blog posts, flashcards, and corporate training decks can help, but CDMP terminology is DAMA-specific. If your workplace uses a different title for a steward, owner, custodian, architect, or council, learn the DAMA language for exam day.
Bottom Line
How to Turn DMBoK Reading Into Exam Decisions
The CDMP Fundamentals exam rarely rewards passive reading. A better method is to convert every knowledge area into a set of exam decisions. For governance, ask who has decision rights and what policy or standard should exist. For modeling, ask which level of model is being requested and whether the issue is semantic, structural, or physical. For quality, ask which dimension is failing, what rule would detect it, and what process would remediate it. For metadata, ask whether the scenario needs business meaning, technical lineage, operational processing evidence, or a repository architecture.
Build a one-page comparison sheet for terms that sound adjacent: data owner versus steward, custodian versus DBA, policy versus standard, reference data versus master data, registry MDM versus consolidation MDM, ETL versus ELT, data lake versus warehouse, retention schedule versus backup, and maturity assessment versus audit. CDMP questions often become easy once the two closest concepts are separated cleanly.
Buy on Amazon