This is not a pure Data Science, Machine Learning, Statistics, SQL, or requirements-focused Business Analysis certification. Each of those disciplines contributes tools that a Business Analyst must understand and apply appropriately, but none dominates the assessment. Candidates are expected to demonstrate practical analytical judgement — connecting data and method to a real business decision — rather than memorising terminology or performing advanced mathematics.
Official exam format
- Number of questions
- 40
- Time allowed
- 60 minutes
- Marks per correct answer
- +4
- Marks per incorrect answer
- −2
- Marks per unanswered question
- 0
- Maximum possible score
- 160
- Minimum passing score
- 120
- Passing percentage
- 75%
- Target difficulty
- 4/5
- Question format
- Scenario-based, single-best-answer, 4 options
SQL (7.5%) and the three statistics-adjacent domains combined (Statistical Foundations, Probability & Uncertainty, Correlation & Regression — 17.5%) together represent 25% of the exam, ensuring technical content informs but does not dominate a certification whose focus is applied business analytics.
Detailed domain specifications
Domain 1 · 5% · 2 questions
Business Analytics Foundations
Establish what Business Analytics is, how it differs from adjacent disciplines, and the end-to-end analytics decision process.
Learning objectives
- Explain the Business Problem → Data → Analysis → Insight → Recommendation → Business Decision process
- Distinguish Business Analytics from Data Science, BI reporting, and requirements-focused Business Analysis
Knowledge
- Roles and boundaries of Business Analytics vs. Data Science/ML vs. BI/reporting
- The analytics maturity progression (descriptive → diagnostic → predictive → prescriptive)
Practical skills
- Identify which stage of the analytics process a given task belongs to
- Recognise when a request needs deeper analysis vs. simple reporting
Key concepts
- Analytics maturity model
- Data-to-decision pipeline
- Analytics vs. reporting
Frameworks
- Descriptive/Diagnostic/Predictive/Prescriptive maturity model
Example situations
- A stakeholder asks for “more data” when what's actually needed is a specific analysis
- Distinguishing a reporting request from an analytical one
Not covered
- Formal data science engineering, MLOps, or advanced ML theory
Expected level: Application — situating a real request within the analytics process.
Domain 2 · 5% · 2 questions
Business Problem Definition & Analytical Thinking
Assess the ability to translate a vague business question into a well-defined, answerable analytical question.
Learning objectives
- Reframe ambiguous stakeholder requests as specific analytical questions
- Identify what evidence would actually answer the business question
Knowledge
- Problem framing
- Distinguishing symptoms from root causes
- Scoping an analysis appropriately
Practical skills
- Convert “sales are down, why?” into a structured, testable analytical question
- Identify missing information needed before analysis can begin
Key concepts
- Business question vs. analytical question
- Problem decomposition
Frameworks
- Structured problem framing (issue tree style reasoning, used conceptually)
Example situations
- A CEO asks “why are we losing customers?” with no further detail
- A stakeholder requests an analysis that would not actually answer their real question
Not covered
- Formal management-consulting frameworks in depth (e.g., full MECE training)
Expected level: Application in ambiguous, underspecified scenarios.
Domain 3 · 2.5% · 1 question
Data Collection & Data Sources
Assess understanding of where business data comes from and how source choice affects analysis.
Learning objectives
- Identify appropriate data sources for a business question
- Recognise limitations of a given data source
Knowledge
- Common business data sources (transactional systems, CRM, product analytics, surveys, third-party data)
- Primary vs. secondary data
- Sampling from operational systems
Practical skills
- Recognise when a chosen data source is insufficient or biased for the question being asked
Key concepts
- Data source fitness-for-purpose
- Primary vs. secondary data
Frameworks
- None formal — assessed via applied judgement
Example situations
- Using only support-ticket data to estimate overall customer satisfaction
Not covered
- Data engineering/ETL pipeline architecture
Expected level: Application — evaluating source suitability in context.
Domain 4 · 5% · 2 questions
Data Quality & Data Governance
Assess the ability to recognise data quality issues and their impact on analysis.
Learning objectives
- Identify common data quality problems
- Understand basic governance concepts (ownership, definitions, access)
Knowledge
- Completeness, accuracy, consistency, timeliness
- Duplicate records
- Inconsistent definitions across teams
- Data lineage basics
Practical skills
- Recognise when a data quality issue would invalidate a conclusion
- Ask the right clarifying questions before trusting a dataset
Key concepts
- Data quality dimensions
- “Garbage in, garbage out”
- Metric definition consistency
Frameworks
- Core data quality dimensions (completeness, accuracy, consistency, timeliness, validity)
Example situations
- Two teams report different “active user” counts due to inconsistent definitions
- Duplicate customer records inflate a count
Not covered
- Formal data governance certification (e.g., DAMA-DMBOK in depth) or master data management tooling
Expected level: Application — spotting a quality issue embedded in a scenario.
Domain 5 · 5% · 2 questions
Data Preparation & Cleaning
Assess understanding of how raw data must be prepared before reliable analysis.
Learning objectives
- Identify necessary cleaning/preparation steps for a given dataset
- Recognise the risk of skipping preparation
Knowledge
- Handling missing values
- Outlier identification
- Data type/format issues
- Deduplication
- Merging/joining datasets from different sources
Practical skills
- Decide an appropriate treatment for missing or outlier data given the business context
- Recognise when a “clean-looking” dataset still has hidden issues
Key concepts
- Missing data handling approaches
- Outlier treatment
- Data transformation
Frameworks
- None formal — applied judgement on preparation trade-offs
Example situations
- A revenue field has a small number of implausibly large values that skew the average
- A dataset merge produces unexpected duplicate rows
Not covered
- Advanced feature engineering for machine learning models
Expected level: Application — choosing appropriate, context-sensitive preparation steps.
Domain 6 · 5% · 2 questions
Exploratory Data Analysis
Assess the ability to explore a dataset to generate initial understanding before formal analysis.
Learning objectives
- Use summary statistics and simple visual exploration to identify patterns, anomalies, and questions worth pursuing
Knowledge
- Summary statistics (mean, median, mode, range)
- Distribution shape
- Identifying anomalies through exploration
Practical skills
- Interpret a simple data summary to decide what deeper analysis is warranted
- Spot a pattern that contradicts an initial assumption
Key concepts
- Exploratory vs. confirmatory analysis
- Distribution shape (skew, spread)
Frameworks
- None formal — exploratory analysis practice
Example situations
- Initial exploration reveals a bimodal distribution in customer spend, suggesting two distinct customer groups worth investigating separately
Not covered
- Advanced statistical exploratory techniques (e.g., formal multivariate exploratory methods)
Expected level: Application — interpreting exploratory summaries to guide next steps.
Domain 7 · 5% · 2 questions
Descriptive Analytics
Assess the ability to summarise “what happened” accurately and communicate it appropriately.
Learning objectives
- Select appropriate summary measures
- Avoid misleading descriptive summaries
Knowledge
- Mean vs. median vs. mode and when each is appropriate
- The effect of outliers on averages
- Aggregation pitfalls
Practical skills
- Recognise when a mean is misleading due to skew and a median would better represent typical experience
- Choose the right level of aggregation for a business question
Key concepts
- Central tendency measures
- The effect of skewed distributions on averages
Frameworks
- None formal — applied descriptive statistics judgement
Example situations
- A rising average order value driven by a handful of large outlier orders, masking flat typical customer spend
Not covered
- Advanced descriptive statistical theory
Expected level: Application — interpreting and critiquing descriptive summaries.
Domain 8 · 7.5% · 3 questions
Diagnostic Analytics
Assess the ability to investigate “why” something happened, beyond simply describing that it happened.
Learning objectives
- Structure an investigation into a metric change
- Distinguish correlation from causation while diagnosing a cause
Knowledge
- Root-cause investigation approaches
- Segmenting data to isolate drivers
- The difference between a symptom and a cause
Practical skills
- Given a metric decline, identify what further breakdowns/segments would help isolate the cause
- Avoid jumping to a causal conclusion prematurely
Key concepts
- Root-cause analysis
- Segmentation for diagnosis
- Confounding factors
Frameworks
- Segment-and-isolate diagnostic approach (conceptual)
Example situations
- Overall conversion rate drops, but the cause is isolated to one traffic channel once segmented
- A metric decline coincides with, but is not necessarily caused by, a recent change
Not covered
- Formal causal inference statistical methods (e.g., instrumental variables, formal experimental design theory)
Expected level: Application — this is a heavily tested, core domain.
Domain 9 · 5% · 2 questions
Statistical Foundations for Business Analytics
Assess practical, applied understanding of core statistical concepts as used in business analysis.
Learning objectives
- Interpret variance and standard deviation in a business context
- Understand percentiles and their business use
Knowledge
- Mean, median, mode, variance, standard deviation, percentiles
- What these measures communicate about spread and typical values
Practical skills
- Interpret a reported standard deviation or percentile in the context of a business decision
- Recognise when spread matters as much as the average
Key concepts
- Measures of central tendency and spread
- Percentile interpretation
Frameworks
- None formal — applied statistical literacy
Example situations
- A high standard deviation in delivery times suggests inconsistent service even if the average delivery time looks acceptable
Not covered
- Formal mathematical statistics or proofs
Expected level: Application/interpretation, not calculation-heavy.
Domain 10 · 5% · 2 questions
Probability & Uncertainty
Assess practical understanding of probability, sampling, and uncertainty as they affect business conclusions.
Learning objectives
- Understand sampling and its effect on confidence in a conclusion
- Interpret confidence intervals and basic probability statements in business terms
Knowledge
- Sampling basics (sample size, sampling bias)
- Confidence intervals (conceptual interpretation)
- Basic probability reasoning
Practical skills
- Recognise when a conclusion is based on too small or biased a sample
- Interpret what a confidence interval does and does not tell you
Key concepts
- Sampling bias
- Confidence interval interpretation
- Uncertainty communication
Frameworks
- None formal — applied probabilistic reasoning
Example situations
- A survey of 20 customers is used to make a company-wide product decision
- A reported confidence interval is misinterpreted as a guarantee
Not covered
- Formal probability theory, distributions mathematics, or Bayesian statistics in depth
Expected level: Application/interpretation — no formal probability calculation required.
Domain 11 · 7.5% · 3 questions
Correlation & Regression
Assess the ability to interpret relationships between variables and avoid common misinterpretations.
Learning objectives
- Distinguish correlation from causation
- Interpret a simple regression coefficient in business terms
- Recognise confounding variables
Knowledge
- Correlation coefficient interpretation (direction/strength, conceptually)
- Regression coefficients as business interpretation, not formula derivation
- Confounding and spurious correlation
Practical skills
- Identify a plausible confounding factor behind an observed correlation
- Interpret what a regression coefficient implies about a business relationship, and its limitations
Key concepts
- Correlation vs. causation
- Confounding variables
- Regression coefficient interpretation
Frameworks
- Simple linear regression (conceptual interpretation, not derivation)
Example situations
- Ice-cream-sales-and-drowning-style spurious correlation scenarios in a business context
- A regression suggesting price affects demand, but a confound (e.g., seasonality) may be responsible
Not covered
- Regression mathematics/formula derivation, multivariate regression theory, or advanced model diagnostics
Expected level: Application — this is a heavily tested, core domain.
Domain 12 · 5% · 2 questions
Predictive Analytics Fundamentals
Assess conceptual understanding of predictive modelling as a business tool, without requiring model-building skill.
Learning objectives
- Understand what a predictive model does and does not tell a business
- Interpret basic model evaluation concepts
Knowledge
- Classification vs. regression prediction tasks (conceptual)
- Overfitting
- Train/test split concept
- Precision/recall (conceptual, business-interpreted)
- Interpreting model outputs and their uncertainty
Practical skills
- Recognise signs of overfitting described in a business scenario
- Interpret what a model's precision/recall trade-off means for a business decision (e.g., fraud detection)
Key concepts
- Overfitting
- Train/test concept
- Precision vs. recall trade-off
- Model interpretability vs. accuracy trade-off
Frameworks
- Train/test evaluation concept
- Precision-recall trade-off (conceptual)
Example situations
- A model performs very well on training data but poorly in production, indicating overfitting
- A fraud model with high recall but low precision creates excessive false positives
Not covered
- Model-building, coding, algorithm selection, or advanced ML techniques
Expected level: Conceptual application — interpreting model behaviour and outputs, not building models.
Domain 13 · 5% · 2 questions
Forecasting & Time-Series Analysis
Assess the ability to interpret forecasts and time-series patterns in a business context.
Learning objectives
- Distinguish trend from seasonality
- Interpret forecast uncertainty and forecast error
Knowledge
- Trend vs. seasonality vs. noise
- Forecast horizon and uncertainty growth over time
- Forecast error interpretation
Practical skills
- Identify seasonality being mistaken for a trend (or vice versa)
- Interpret why a forecast's uncertainty widens further into the future
Key concepts
- Trend, seasonality, noise
- Forecast error
- Forecast horizon
Frameworks
- Trend/seasonality/noise decomposition (conceptual)
Example situations
- A retailer mistakes predictable holiday seasonality for a genuine growth trend
- A long-range forecast is treated as equally certain as a near-term one
Not covered
- Formal time-series modelling techniques (ARIMA, exponential smoothing mathematics) or coding forecasts
Expected level: Application/interpretation — no formal time-series modelling required.
Domain 14 · 7.5% · 3 questions
Business Metrics & KPI Design
Assess the ability to select, define, and critique business metrics and KPIs.
Learning objectives
- Design a KPI that reflects genuine business value
- Recognise vanity metrics and metric-gaming risk
Knowledge
- Leading vs. lagging indicators
- Vanity metrics
- Metric definitions and consistency
- North Star-style unifying metrics
Practical skills
- Critique a proposed KPI for whether it reflects real business value or is easily gamed
- Select an appropriate metric for a given business goal
Key concepts
- Leading/lagging indicators
- Vanity metrics
- Metric gaming
Frameworks
- North Star metric concept (conceptual)
- Leading/lagging indicator framing
Example situations
- A team optimises a metric that improves on paper while the underlying business outcome worsens
- Choosing between two candidate KPIs for a new product goal
Not covered
- Formal balanced-scorecard certification content
Expected level: Application — this is a heavily tested, core domain.
Domain 15 · 5% · 2 questions
Data Visualisation
Assess the ability to choose appropriate visualisations and avoid misleading charts.
Learning objectives
- Select the right chart type for a given data story
- Identify misleading visualisation practices
Knowledge
- Chart type selection (bar, line, scatter, etc.) for different data relationships
- Common distortions (truncated axes, inappropriate scales, misleading dual axes)
Practical skills
- Identify why a given chart misrepresents the underlying data
- Choose an appropriate chart for a specific comparison or trend
Key concepts
- Chart-type fitness for purpose
- Visual distortion
- Communicating uncertainty visually
Frameworks
- Chart-selection-by-purpose guidance (conceptual, not a rigid rulebook)
Example situations
- A truncated y-axis exaggerates a small change
- A line chart is used for unrelated categorical data
Not covered
- Specific visualisation software mechanics (e.g., a particular tool's menu steps)
Expected level: Application — critiquing and selecting visualisations in context.
Domain 16 · 2.5% · 1 question
Dashboard Design & Reporting
Assess the ability to design dashboards and reports that serve real business decisions.
Learning objectives
- Design a dashboard around the audience's actual decisions
- Avoid dashboard clutter and metric overload
Knowledge
- Audience-appropriate reporting
- The difference between an operational and executive dashboard
- Avoiding excessive metric density
Practical skills
- Critique a dashboard for whether it serves its intended audience and decision
- Simplify an overloaded report
Key concepts
- Audience-first dashboard design
- Signal vs. noise in reporting
Frameworks
- None formal — applied dashboard design judgement
Example situations
- An executive dashboard contains 40 metrics with no clear priority or story
Not covered
- Specific BI tool configuration steps
Expected level: Application — evaluating and improving dashboard/report design.
Domain 17 · 7.5% · 3 questions
SQL & Data Querying Concepts
Assess practical understanding of SQL as an analytical tool, not database administration.
Learning objectives
- Reason about SELECT, WHERE, GROUP BY, ORDER BY, JOIN, aggregation, CASE logic, subqueries, and window-function concepts in a business analysis context
Knowledge
- Filtering and aggregating data
- Joining tables conceptually (inner/left join effects on row counts)
- Conditional (CASE) logic
- The purpose of subqueries
- The conceptual purpose of window functions (e.g., running totals, rankings) without requiring exact syntax recall
Practical skills
- Determine what a given query would return, or what query logic would answer a business question, from a described schema/scenario
Key concepts
- Filtering vs. aggregating
- Join effects on result rows
- Window-function use cases
Frameworks
- None formal — applied SQL reasoning over described schemas
Example situations
- Determining why a join produced more rows than expected (fan-out)
- Choosing GROUP BY logic to answer a specific aggregation question
Not covered
- Database-specific syntax variations, query performance tuning, or database administration
Expected level: Application — reasoning about query logic and outcomes, not writing exact syntax from memory.
Domain 18 · 5% · 2 questions
Analytical Decision-Making
Assess the ability to translate analysis into a sound business decision or recommendation.
Learning objectives
- Determine whether evidence sufficiently supports a proposed decision
- Identify what additional analysis is needed before deciding
Knowledge
- Evidence sufficiency
- The risk of acting on a single metric
- Trade-offs between analytical rigor and decision timeliness
Practical skills
- Given a metric change, determine what further investigation is needed before recommending action
- Recognise when a decision is being made prematurely
Key concepts
- Evidence-based decision-making
- Single-metric risk
- Decision timeliness vs. rigor trade-off
Frameworks
- None formal — applied decision judgement
Example situations
- A team wants to act immediately on a conversion drop without first checking for a data or tracking issue
Not covered
- Formal decision-theory mathematics (e.g., expected utility calculations)
Expected level: Application — this is a heavily tested, core domain connecting analysis to action.
Domain 19 · 2.5% · 1 question
Communicating Insights & Recommendations
Assess the ability to communicate analytical findings clearly and honestly to a business audience.
Learning objectives
- Structure a recommendation that connects data to a clear business action
- Communicate uncertainty and limitations honestly
Knowledge
- Insight communication structure (finding → implication → recommendation)
- Avoiding overstatement of certainty
Practical skills
- Rewrite an overstated or jargon-heavy finding into a clear, appropriately-caveated business recommendation
Key concepts
- Data storytelling
- Communicating uncertainty
- Audience-appropriate framing
Frameworks
- Finding-implication-recommendation structure (conceptual)
Example situations
- An analyst presents a finding as certain when the underlying evidence is preliminary
Not covered
- Formal public-speaking or presentation-design training
Expected level: Application — evaluating and improving communication of findings.
Domain 20 · 2.5% · 1 question
Ethics, Bias & Responsible Analytics
Assess the ability to recognise ethical risk in analytical work.
Learning objectives
- Identify bias in data or analysis
- Recognise data privacy and responsible-use considerations
Knowledge
- Sampling and selection bias
- Algorithmic/data bias affecting fairness
- Data privacy basics
- The risk of misleading conclusions (intentional or not)
Practical skills
- Identify a biased sample or a potentially unfair use of a model/analysis
- Recognise when a conclusion is being presented misleadingly
Key concepts
- Selection bias
- Fairness
- Data privacy
- Responsible use of analytics
Frameworks
- None formal — applied ethical judgement
Example situations
- A model trained on historically biased data perpetuates unfair outcomes
- A chart is constructed in a way likely to mislead a non-technical audience
Not covered
- Formal legal/regulatory certification (e.g., GDPR/CCPA compliance qualification)
Expected level: Application — recognising ethical risk embedded in a normal analytics scenario.
Competency map
- Foundational Analytical Thinking
- Domains 1, 2
- Data Readiness
- Domains 3, 4, 5
- Exploration & Description
- Domains 6, 7, 8
- Statistical & Quantitative Reasoning
- Domains 9, 10, 11
- Forward-Looking Analytics
- Domains 12, 13
- Measurement & Communication
- Domains 14, 15, 16, 19
- Technical Querying
- Domains 17
- Decision & Responsibility
- Domains 18, 20
