- The Short Answer: What CDS Actually Is
- Who Issues the Credential
- The 12 Topics the Exam Covers
- What Makes It Different From a Tool-Based Certification
- Fees, Proctoring and Application Mechanics
- Keeping the Credential Active
- Who Should Pursue It and Who Hires for It
- A Domain-Ordered Preparation Sequence
- What Is Not Publicly Confirmed
- Frequently Asked Questions
- CDS here means Certified Data Scientist - Associate/Practitioner, issued by the Institute for the Certification of Computing Professionals (ICCP).
- The accessible Data Science Exam outline lists 12 topics, from business questions through reproducibility, without published percentage weights.
- ICCP's store lists professional exams at $350 including proctoring; the generic application fee is $45.
- Certification runs on a 3-year professional-development cycle; Associate and Practitioner tiers differ in required hours and annual fees.
The Short Answer: What Actually Is CDS?
CDS, in the context of this site, stands for Certified Data Scientist - Associate/Practitioner. It is a professional credential for people who work with data to answer questions, build models and communicate results. The acronym is shared by several unrelated credentials in other industries, so it is worth being precise: everything on this page concerns the data science certification issued through ICCP, not any other certification that happens to abbreviate to the same three letters.
The credential is offered at an Associate level and a Practitioner level. That two-tier framing matters because it signals who the certification is aimed at: people early in a data science career (Associate) and people with more established practice (Practitioner). If you want the terminology unpacked further, our pages on what CDS stands for and what CDS certification is cover the naming in more depth.
Who Issues the Credential
The certifying body is the Institute for the Certification of Computing Professionals (ICCP). ICCP is a long-running certifying organization for computing professionals, and its credentials are structured around examinations, an ethics commitment and ongoing professional development rather than a one-time test you pass and forget.
Two features of the ICCP model shape what the CDS is like in practice:
- Examinations are proctored. ICCP's general exam process uses approved proctoring, with arranged remote or in-person options. The specific testing vendor and workflow for the CDS exam have not been verified publicly, so confirm the details when you register.
- Certification is a professional commitment. The application includes a resume and an ethics agreement, and holding the credential means completing professional-development activity on a recurring cycle.
The 12 Topics the Exam Covers
The currently accessible Data Science Exam outline lists 12 official topics. The outline does not publish percentage weights or a revision year, and it should not be treated as a verified blueprint for a complete multi-exam pathway. Within those limits, it tells you a great deal about the philosophy of the credential: it begins with the business problem and ends with reproducibility, which is the full arc of a real analytics project rather than a narrow slice of it.
For a deeper treatment of each area, see our complete guide to all 12 CDS content areas. Here is the shape of the outline.
Framing and Foundations
Domain 1: Business & Technology Issues: Starting with the Question First, What Problem Are You Trying to Solve?
The outline opens with problem definition, not algorithms. Expect to reason about what a stakeholder actually needs before any data is touched.
- Translating a vague business ask into an answerable data question
- Recognizing when data science is, and is not, the right tool
- Understanding technology constraints that shape a solution
Domain 2: Data Storage, Big Data and Sources
Where data lives and how it arrives determines what you can do with it.
- Storage approaches and the trade-offs between them
- Big data concepts and when scale changes the approach
- Evaluating the origin and reliability of data sources
Domain 3: Mathematics and Statistical Data Science
The quantitative backbone of the discipline. This domain is the reason candidates with purely tool-driven backgrounds often find the credential more demanding than expected.
- Core mathematical concepts underlying analysis
- Probability and statistical reasoning
Domain 4: Programming Skills
Programming appears as one topic among twelve, not the centerpiece. Expect it to support the analytical work rather than dominate it.
- Programming concepts relevant to data work
- Working with data programmatically
Working With Data
Domain 5: The Data Analytic Question and Types of Data and Reporting
Different questions call for different analytic approaches, and different data types call for different treatment.
- Matching the type of question to the type of analysis
- Understanding data types and how they affect method selection
- Reporting results appropriately
Domain 6: Tidying the Data - Data Cleaning and Quality
Real data is messy. This domain tests whether you understand how to make it usable and how to judge its quality.
- Identifying and handling quality problems
- Preparing data into an analysis-ready form
Domain 7: Exploratory Analysis
Looking before modeling: summarizing, visualizing and probing data to understand its structure and surprises.
- Descriptive summaries and visual inspection
- Forming hypotheses from patterns in the data
Modeling and Interpretation
Domain 8: Statistical Modeling and Inference
Moving from describing a sample to drawing conclusions about a population, with appropriate caution about uncertainty.
- Fitting and interpreting statistical models
- Reasoning about inference and its limits
Domain 9: Prediction and Machine Learning
Building models whose goal is to predict well, and evaluating whether they do.
- Machine learning concepts and approaches
- Assessing predictive performance honestly
Domain 10: Causality
Distinguishing correlation from cause. Few certifications give causality its own topic, and its presence here reflects a rigorous view of what data science is for.
- What it takes to support a causal claim
- Why prediction and causal explanation are different goals
Communication and Rigor
Domain 11: Written Analysis
An analysis nobody can understand has no impact. This topic treats clear written communication of findings as a core professional skill.
- Structuring and explaining an analysis in writing
- Tailoring the message to the audience
Domain 12: Reproducibility
The outline closes with the ability to repeat and verify an analysis.
- Why reproducible work matters for trust
- Practices that let others rerun and check your results
What Makes It Different From a Tool-Based Certification
Many candidates arrive expecting a credential about a particular language or platform. The 12-topic outline points somewhere else. Notice how the list is distributed:
| Concern | Where it appears in the outline |
|---|---|
| Problem framing and business context | Domains 1 and 5 |
| Data infrastructure and quality | Domains 2 and 6 |
| Quantitative and statistical reasoning | Domains 3, 8 and 10 |
| Hands-on technical skill | Domain 4, plus modeling in Domain 9 |
| Communication and professional practice | Domains 11 and 12 |
Programming is a single topic, while framing, statistics, causality, writing and reproducibility together occupy most of the list. That balance rewards candidates who think like analysts, not only candidates who can write code. It also means that strong coders sometimes underestimate the exam: fluency in a library does not substitute for understanding inference or causality. For a frank look at where candidates struggle, read how hard the CDS exam is.
Fees, Proctoring and Application Mechanics
Costs and process are where careful reading pays off, because several figures are published but their applicability to your specific CDS tier is not fully confirmed.
What Is Published
- Professional exam fee: ICCP's current store lists $350 per professional exam, including proctoring. An older FAQ listed $299, so rely on the current store listing and confirm at purchase.
- Foundational exam fee: The store lists $249 per foundational exam.
- Application fee: The generic candidate application fee is $45.
- Application contents: The general application requires a resume and an ethics agreement.
What Is Not Confirmed
It is not verified which exam products the CDS tier you want requires, or how many exams that tier involves. For that reason there is no responsible way to state a total certification-exam cost here. Add up the exact exams ICCP requires for your tier, plus the application fee, after confirming with the issuer. Our CDS certification cost breakdown walks through how to build that budget, and CDS requirements covers the eligibility questions.
| Item | Published figure | Status |
|---|---|---|
| Professional exam (with proctoring) | $350 | Current store listing |
| Foundational exam | $249 | Current store listing |
| Candidate application | $45 | Generic ICCP fee |
| Total for your tier | Not stated | Depends on required exams; confirm with ICCP |
Keeping the Credential Active
Passing is not the end of the commitment. An active certification follows a 3-year professional-development cycle. ICCP's general tier rules specify:
- Associate: 60 hours of professional development and a $35 annual fee.
- Practitioner: 120 hours of professional development and a $75 annual fee.
Whether these exact figures apply to your individual circumstances should be confirmed with ICCP. Even so, the structure tells you something useful: this is a credential that expects continued learning, which is part of why employers who care about professional standards take ICCP certifications seriously.
Who Should Pursue It and Who Hires for It
The credential is a natural fit for several kinds of candidates:
- Analysts moving toward data science who want a structured, recognized way to demonstrate breadth beyond spreadsheets and dashboards.
- Self-taught practitioners whose knowledge is deep in places and patchy in others; the 12-topic outline exposes the gaps.
- Software or database professionals adding statistical and causal reasoning to a strong technical base.
- Working data scientists who want an independent professional credential tied to an ethics commitment.
On the hiring side, the roles that tend to value a credential like this are analytics, data science and data-driven decision-making positions in organizations that care about professional standards, documentation and defensible methodology. The emphasis on written analysis and reproducibility is particularly relevant to regulated or audit-conscious environments. No specific salary uplift is verified here, so treat earnings claims with caution and see our CDS salary guide and CDS jobs page for how to evaluate the market. For a structured view of the return on investment, read whether the CDS certification is worth it.
Key Takeaway
If your experience is strongest in programming, spend deliberate time on statistical modeling, causality and written analysis. If your background is statistics or research, invest in data storage, programming and reproducibility. The outline rewards balance.
A Domain-Ordered Preparation Sequence
Rather than a generic schedule, order your study to follow the logic of the outline, since later topics depend on earlier ones. Here is one sensible arrangement; stretch or compress it to your timeline, and see the full CDS study guide for more detail.
Framing and data foundations
- Domain 1: practice turning vague requests into precise data questions
- Domain 2: review storage types, big data concepts and source evaluation
Quantitative core
- Domain 3: mathematics and statistical fundamentals, the hardest area to cram later
- Domain 4: programming practice applied to small data tasks
Preparing and exploring data
- Domain 5: question types, data types and reporting
- Domains 6 and 7: cleaning, quality checks and exploratory analysis on a messy dataset
Modeling and causal thinking
- Domains 8 and 9: inference versus prediction, and how each is evaluated
- Domain 10: practice explaining why a correlation does not establish a cause
Communication, reproducibility and review
- Domains 11 and 12: write up one analysis and make it rerunnable
- Take timed practice questions across all 12 topics and revisit weak areas
Because the outline does not publish topic weights, avoid over-investing in a single area on the assumption that it counts for more. Spread effort across all twelve, and use realistic practice questions on the CDS Exam Prep practice test site to find where you are weakest. A one-page recap such as the CDS cheat sheet is useful in the final days.
What Is Not Publicly Confirmed
Honest preparation means knowing which facts you still need to verify. The following are not confirmed in the information available for this credential:
- The exact education, experience, training-hour and reference prerequisites for the CDS tier you intend to pursue.
- Which exams, and how many, the requested tier requires.
- The CDS-specific testing vendor and workflow.
- Passing score, pass rate, question count and time limit.
- Open-book status, calculator rules and whether the exam adapts to your answers.
- Topic percentage weights and the revision year of the outline.
Treat any source that states these with great precision and no citation skeptically. Resources like our pages on the CDS passing score, CDS pass rate and CDS exam dates explain what is and is not known, so you can plan around real information. When in doubt, ask ICCP directly before paying any fee.
Frequently Asked Questions
CDS stands for Certified Data Scientist - Associate/Practitioner, a credential issued by the Institute for the Certification of Computing Professionals (ICCP). Other certifications share the abbreviation, but this site covers only the ICCP data science credential.
The currently accessible Data Science Exam outline lists 12 official topics, running from business problem framing through reproducibility. It does not publish percentage weights or a revision year, so it should not be treated as a complete verified blueprint.
ICCP's store lists $350 per professional exam including proctoring and $249 per foundational exam, with a generic $45 candidate application fee. Which exams your tier requires is not verified, so confirm the total with ICCP before budgeting.
Active certification follows a 3-year professional-development cycle. ICCP's general tier rules list 60 hours and a $35 annual fee for Associate, and 120 hours and a $75 annual fee for Practitioner; confirm how these apply to you.
No. Programming is one of 12 topics. The outline gives substantial space to problem framing, statistics, causality, written analysis and reproducibility, so a purely coding-focused preparation leaves large gaps.