CBDA Mock Questions: 35 Practice Questions Matching the Official Exam Blueprint
Aliaksei Khavanski
Expert Contributor
June 12, 2026
Last Updated
How do you know you are ready for the CBDA? You take a mock that behaves like the real thing. This set of 35 practice questions mirrors the official IIBA exam blueprint, so each of the six domains carries the same relative weight you will face on exam day: Identify the Research Question (7 questions, ~20%), Source Data (5, ~15%), Analyze Data (5, ~16%), Interpret and Report Results (7, ~20%), Use Results to Influence Business Decision Making (7, ~20%), and Guide Organization-level Strategy for Business Data Analytics (4, ~9%).
Every question, correct answer, and explanation is grounded in the Guide to Business Data Analytics - each explanation in the answer key cites the exact Guide section, so you can go straight to the source for anything you miss.
How to use this mock: give yourself about 45 minutes, answer all 35 questions without peeking, then score yourself against the answer key at the end. On the real exam you need roughly 70%+, so aim for 25 or more correct. Pay attention to which domain your misses cluster in - that tells you where to focus next.
Ready? Timer on.
Identify the Research Question (7 questions)
Question 1. BrightBasket, a retail chain, has just funded an analytics initiative and there is strong pressure to show quick results. A junior analyst proposes immediately building a forecasting model. According to the Guide, what failure mode is the team most at risk of, and what should be done first?
A. The team risks the jump to solution focus; it should first define the business problem or opportunity with the right stakeholders
B. The team risks slow delivery; it should first buy a faster analytics platform before defining anything
C. The team risks the jump to solution focus; it should first pick the predictive algorithm and tune it
D. The team risks weak reporting; it should first design the final dashboard layout for executives
Question 2. At Meridian Bank, the analytics team is identifying stakeholders for a new initiative. The lead analyst argues that all stakeholder groups can be treated the same because they all want the same results. Why does the Guide consider this assumption wrong?
A. Because each stakeholder group articulates different needs, poses different research questions, and holds different skills for interpreting results
B. Because stakeholders should be excluded until the analysis is finished to keep the results objective
C. Because only the sponsor matters and every other stakeholder view can be safely ignored
D. Because all stakeholder groups want identical volumes and timings of analytics results in every case
Question 3. A hospital network wants to reduce patient readmissions and asks an analyst to assess the current state before defining changes. According to the Guide, what is the primary purpose of this current state assessment?
A. To set a baseline and context so the results of data analysis can be better interpreted and improvement can be measured
B. To select the final statistical model that the data scientist will run on the readmission data
C. To approve the data sources and define the acceptance criteria for data accuracy
D. To package the findings into an executive report with the main themes highlighted
Question 4. Nordwind Manufacturing is defining the future state for an analytics initiative aimed at reducing scrap rates. The plant manager states a broad ambition to make things better. According to the Guide on defining the future state, what should the analyst ensure is in place?
A. Measurable objectives are established and key stakeholders share a vision developed by consensus of the outcome sought
B. The exact regression coefficients are fixed before any data is collected from the production line
C. A single sponsor decides the desired outcome alone, without consulting other stakeholders
D. The data is cleansed and joined into a single coherent dataset ready for modelling
Question 5. A telecom operator has a business need to improve customer experience. An analyst drafts the question: Will customer experience improve by adding a pay wallet feature? Using the analytics types described in the Guide, which type of analytics does this question call for?
A. Prescriptive analytics, because it asks whether a specific action will produce a desired improvement
B. Descriptive analytics, because it only summarises what the factors influencing experience are
C. Diagnostic analytics, because it explains why past experience scores changed
D. Predictive analytics, because it only classifies a transaction as a positive or negative experience
Question 6. A city transport agency faces an analytics initiative whose solution approach is more complex than first anticipated. According to the Guide on planning the analytics approach, how are such initiatives typically implemented across stages of maturity?
A. Through a proof of concept for feasibility, then a pilot at limited scale, then a production stage focused on business value
B. Through a single production rollout first, deferring any feasibility or pilot work until after launch
C. Through one fixed plan that cannot change once the scope and deliverables are agreed
D. Through skipping the proof of concept whenever the budget is tight, going straight to full production
Question 7. A SaaS analytics lead wants a technique to keep the team focused on the most urgent items while formulating research questions, sharing insights, and recommending actions. Which technique does the Guide list for this purpose in the Identify the Research Question domain?
A. Prioritization, used throughout the effort to focus attention on the most urgent items
B. Data preprocessing, used to scale and normalise variables before modelling
C. Boxplot analysis, used to read Q1, median, Q3, and outliers from a distribution
D. Functional decomposition, used to break high-level tasks into lower-level activities
Source Data (5 questions)
Question 8. A freight logistics firm is planning data collection. Some records sit in a SQL data warehouse queried with structured queries, while driver feedback arrives as free-text emails and social media posts. According to the Guide, how should the analyst classify these two kinds of data?
A. The warehouse records are structured data and the free-text feedback is unstructured data
B. The warehouse records are unstructured data and the free-text feedback is structured data
C. Both kinds are structured data because they are both stored digitally by the firm
D. Both kinds are unstructured data because they both required collection over business processes
Question 9. An energy utility is determining the data sets for an outage-prediction initiative. A sensor feed shows uncertainties and inconsistencies that make its trustworthiness questionable. Using the Guide terms for data characteristics, which characteristic is most directly in question here?
A. Veracity, the trustworthiness of the data and that which presents uncertainties and inconsistencies
B. Value, the necessity of driving the analytics exercise from real, valuable business goals
C. Volume, the sheer quantity of records the sensor feed produces each day
D. Velocity, the speed at which the sensor feed delivers new readings
Question 10. A retailer is collecting data for a loyalty study. Point-of-sale transaction logs already exist from everyday purchases, but the team also fields a new Likert-scale survey to ask shoppers about satisfaction. According to the Guide, how should these two collection approaches be classified?
A. The point-of-sale logs are passive data collection and the survey is active data collection
B. The point-of-sale logs are active data collection and the survey is passive data collection
C. Both are passive data collection because the retailer owns the customer relationship
D. Both are active data collection because the retailer chose to study loyalty deliberately
Question 11. During data validation at an insurance arm of a bank, an analyst finds that several required policy fields contain null values. Using the Guide data quality characteristics, which characteristic is most directly violated, and how is it assessed?
A. Completeness, assessed by ensuring required fields do not include null values
B. Accuracy, assessed by comparing numbers in a front-end system with the database
C. Consistency, assessed by how reliable the data is across different stores
D. Veracity, assessed by confirming the data drives a valuable business goal
Question 12. A healthcare analytics team is validating data and needs to define how fields in two source systems correspond to fields in the target system. Which Source Data technique does the Guide describe for creating this source-to-target correspondence?
A. Data mapping, used to create a source-to-target data map between data sources and the target system
B. Business rules analysis, used to understand the rules governing what should be validated
C. Functional decomposition, used to break high-level tasks into lower-level activities
D. Root cause analysis, used to trace a reported defect back to its underlying cause
Analyze Data (5 questions)
Question 13. At Steinmetz Components, a business analyst drafts an initial data analysis plan and hands it to the data scientist for review. According to the Guide, who ultimately decides how the data analysis will be conducted, and why?
A. The data scientist decides, because they possess the deep technical expertise on how the analysis is conducted
B. The business analyst decides alone, because they wrote the initial draft of the plan
C. The sponsor decides, because they fund the initiative and own the budget
D. The reporting team decides, because they will present the final results to stakeholders
Question 14. A telecom analytics team is preparing data by joining a subscribers table to a usage table. The data scientist examines whether each subscriber maps to one usage record or to many. According to the Guide on preparing data, what is the team establishing here?
A. The relationships between data, such as whether tables have a 0 to 1, 1 to 1, or 1 to many relationship
B. The trustworthiness of the data that presents uncertainties and inconsistencies
C. The communication needs and preferred channels of each stakeholder group
D. The ranking of solution options before proposing a recommendation
Question 15. An analyst at a public benefits agency runs an initial exploratory analysis on a newly collected survey dataset before any formal modelling. According to the Guide, what is the main purpose of this Explore Data step?
A. To perform a quality check that confirms the right type and quality of data is being obtained before detailed analysis
B. To produce the final stakeholder report with the main themes and a narrative
C. To approve the data sources with business stakeholders and set acceptance criteria
D. To rank candidate solution options and propose a recommendation to decision-makers
Question 16. A SaaS pricing team performs data analysis and observes that when a predictor variable changes, the price changes by a proportional amount along a straight line. Based on the Guide, how should the analyst describe the relationship between this predictor and price?
A. A linear relationship, where a change in the predictor changes the price by a proportional amount
B. A categorical relationship, where the predictor only takes a few unordered labels
C. A relationship with no association, where the predictor does not move with price
D. A bimodal relationship, where the price clusters around two separate peaks
Question 17. A logistics analytics team finds that data exploration passed quality checks, yet the data analysis results still fail to answer the business question. According to the Guide on assessing the analytics approach, what should the team do?
A. Repeat the data exploration and data analysis tasks iteratively until the data and results are acceptable
B. Proceed to recommend actions anyway, since the exploration step already passed its checks
C. Stop the initiative permanently, because results that fail once cannot be improved
D. Hand the unanswered question to the reporting team to present as the final conclusion
Interpret and Report Results (7 questions)
Question 18. While validating its understanding of stakeholders, an energy retailer's analytics team re-checks how quickly results are expected. One executive group's stated turnaround need fell from 20 working days to 15 working days between two reviews. What is the percentage decrease in the stated turnaround need?
A. 25 percent
B. 5 percent
C. 33 percent
D. 75 percent
Question 19. A retail analytics lead is planning stakeholder communication for an iterative initiative. A colleague insists no results should be shared until the very end. According to the Guide on planning stakeholder communication, why is this stance mistaken?
A. Because communications can involve an intermediate or final result, since analytics initiatives are inherently iterative
B. Because every stakeholder must receive the identical level of detail regardless of expertise
C. Because confidentiality is irrelevant once an initiative has formally started
D. Because feedback from stakeholders should be discarded rather than recorded for follow-up
Question 20. A bank's analytics team is determining the communication needs of different stakeholder groups so each message is clearly understood. According to the Guide, which of the following is a stakeholder communication preference the team should capture?
A. What information is most relevant to them and how often they wish to be updated
B. The exact hyperparameters the data scientist will use in the model
C. The source-to-target data map between the data sources and the target system
D. The five named data quality characteristics for technical validation
Question 21. A hospital analytics team merges two datasets and finds a striking correlation between two variables. An executive wants to build a prediction model that treats one variable as the cause of the other. According to the Guide on deriving insights, what is the correct stance?
A. Treat it as a correlation, since a fascinating relationship should not be modelled as causal without stronger evidence
B. Treat the correlation as proof of causation, because the relationship is strong and striking
C. Discard both datasets, because any correlation between merged sources is meaningless
D. Report the correlation as a final business insight without any further analysis
Question 22. A manufacturing analyst must report a single headline metric, the overall scrap-rate percentage, to executives. Following the Guide visualization best practices, what is the most effective way to communicate this single metric?
A. Use simple text, since a single or couple of metrics are often communicated more effectively as text
B. Use a complex multi-series chart so the single metric looks more visually impressive
C. Use a pie chart so the audience can interpret the arc lengths and angles
D. Use a three-dimensional column so the single metric stands out from the page
Question 23. A telecom analyst shows an S-curve between marketing expense and revenue and must state, in one succinct line, that beyond a certain spend marketing no longer lifts revenue. Which Interpret and Report Results technique does the Guide describe for communicating this single most relevant finding?
A. The Big Idea, used to communicate the most relevant findings in a succinct manner and answer the so-what question
B. Data mapping, used to define a source-to-target map between data sources and a target system
C. Functional decomposition, used to break high-level tasks into lower-level activities
D. Data profiling, used to assess data quality before the analysis begins
Question 24. In the Guide case study, a premium retailer wanted to measure the success of its social media marketing for a new mix-ins offering. The data science team studied regional adoption and reported which channel had the highest adoption to recommend it as the optimal medium. Which channel did they identify?
A. Facebook, which had the highest regional adoption and was conducive to the research problem
B. A printed in-store flyer, since regional adoption favours offline media
C. Email newsletters, because they reach the widest national audience
D. Radio advertising, because it had the highest regional adoption in the study
Use Results to Influence Business Decision Making (7 questions)
Question 25. A SaaS analytics team has produced valuable insights and now drives conversations about how changes will be made. They list several possibilities, rate and rank them, and propose one. According to the Guide, what are these possibilities called, and what may they include?
A. Solution options, which may include process, tool, resource, or IT system changes
B. Research questions, which are derived from the business need before analysis begins
C. Data quality characteristics, which describe accuracy, completeness, and consistency
D. Communication preferences, which describe how stakeholders wish to be updated
Question 26. A logistics firm has an approved recommendation and must plan its execution. An analyst needs to drill the high-level rollout down into lower-level tasks and activities. According to the Guide on developing the implementation plan, which technique is used for this breakdown?
A. Functional decomposition, used to drill high-level tasks down into lower-level tasks and activities
B. Likert-scale survey design, used to actively collect self-reported data from customers
C. Data cleansing, used to correct or remove bad data before analysis
D. Boxplot interpretation, used to read Q1, median, Q3, and outliers from a distribution
Question 27. At an energy provider, analysis results must become implemented policies and procedures. A business analyst takes the change manager role to oversee this transformation. According to the Guide, why are business analysts well suited to this role?
A. Because they ensure continuity between the analytics work and its implementation
B. Because they alone hold the deep technical expertise to build the predictive model
C. Because they replace stakeholder agreement with their own judgement on changes
D. Because they finalise the data quality acceptance criteria for the sources
Question 28. A retailer wants to evaluate how competitors and the market might react if it implements a recommendation from its analytics initiative. Which technique does the Guide list for the Use Results to Influence Business Decision Making domain for this purpose?
A. Benchmarking and Market Analysis, used to evaluate competitor and market reactions to recommendations
B. Data profiling, used to assess data quality before analysis begins
C. Exploratory data analysis, used to surface initial trends before formal modelling
D. Data mapping, used to define a source-to-target map between data and a target system
Question 29. In the Guide case study, retailers in the UK giftware and online gifting industry faced cost pressures and a crowded, historically seasonal marketplace. According to the case, how did several top retailers choose to differentiate themselves?
A. By investing in more customized gift offerings personalized for newer target market segments
B. By withdrawing from online channels to focus only on physical stores
C. By lowering prices uniformly across every product to win on cost alone
D. By halting all marketing spend until the seasonal peak had passed
Question 30. A healthcare analytics team is ready to recommend changes. Before recommending, the Guide says an evaluation is conducted. According to the Guide, what must this evaluation determine, and what happens if no feasible solution is found?
A. Whether the outcome answered the research question; if no feasible solution exists, the cycle repeats with a new research question
B. Whether the dashboard colours are correct; if not, the report is reformatted and reissued
C. Whether the data sources were approved; if not, the recommendation is published anyway
D. Whether stakeholders liked the visuals; if not, the model is retrained on more data
Question 31. A manufacturer is developing the implementation plan for an approved recommendation. Beyond sequencing the work and showing task dependencies, what else does the Guide say analysts identify and discuss when developing this plan?
A. Constraints, assumptions, risks, and dependencies for implementing the proposed changes
B. The trustworthiness characteristic veracity of each incoming data source
C. The Likert-scale wording for a customer satisfaction survey
D. The exact statistical test the data scientist will run on the dataset
Guide Organization-level Strategy for Business Data Analytics (4 questions)
Question 32. A large telecom enterprise runs many parallel analytics engagements and wants its analytics team structured for standardized practice across business units. According to the Guide on organizational models, which model has an analytics team operating as a single unit, for example as a Centre of Excellence?
A. The centralized model, where the analytics team operates as a single unit supporting other business units
B. The decentralized model, where analytics is embedded separately into each business unit
C. The hybrid model, defined only as analytics embedded into each business unit with no shared unit
D. The proof of concept model, defined as a permanent organizational structure for analytics
Question 33. A public-sector agency is building its analytics talent strategy. Beyond recruiting and retaining people, the Guide names three major components that form the pillars of a robust talent strategy. Which of the following is one of them?
A. Establishing the right team structure for analytics initiatives
B. Selecting the final predictive model for each individual initiative
C. Approving the data sources and setting data acceptance criteria
D. Writing the source-to-target data map for the target system
Question 34. A SaaS company is formulating an organization-level data strategy and wants the planning consideration that covers the rules and policies managing its data assets to ensure high-quality data. Which consideration does the Guide describe for this?
A. Data governance, the rules and policies that manage the data assets to ensure high-quality data
B. Data architecture, the models and standards governing how data is collected and stored
C. Data security, the activities performed to protect data for privacy and confidentiality
D. Metadata management, the administration of information maintained about the data assets
Question 35. A logistics enterprise wants a technique that gives a balanced view of the organization from different perspectives and helps align its data strategy to business objectives and outcomes. Which technique does the Guide list for the Guide Organization-level Strategy domain for this purpose?
A. Balanced Scorecard, used to describe a balanced view of the organization from different perspectives
B. Boxplot, used to read Q1, median, Q3, and outliers from a distribution
C. Data cleansing, used to correct or remove bad data before analysis
D. Likert-scale survey, used to actively collect self-reported customer data
Answer Key and Explanations
1. Correct answer: A. Urgency to see results creates a tendency to jump into a solution focus rather than devote attention to identifying the problem and engaging the right stakeholders, which can lead to a misdiagnosis of the business problem; defining the business problem or opportunity is often the first step. Why others wrong: B - buying a platform is still a solution focus and skips problem definition; C - picking and tuning an algorithm is the jump to solution the Guide warns against; D - designing a dashboard is a reporting activity that presupposes the problem is already defined. Source: Guide Section 2.1.1
2. Correct answer: A. Understanding stakeholders matters because each group articulates different needs and objectives, poses different types of research questions, is interested in different volumes and timings of results, and holds different skillsets and levels of experience with analytics. Why others wrong: B - the Guide says analysts engage and collaborate with stakeholders before the initiative starts, not exclude them; C - the Guide stresses hearing the variety of stakeholders, not only one; D - this contradicts the Guide, which says groups want different volumes and timings of results. Source: Guide Section 2.1.2
3. Correct answer: A. Analyzing the current state sets a baseline and context for making a change, providing context so that results of data analysis can be better interpreted; understanding the current state is fundamental to informed decision-making. Why others wrong: B - selecting a statistical model is an Analyze Data activity, not current state assessment; C - approving data sources and acceptance criteria is data validation in Source Data; D - packaging findings into a report belongs to Interpret and Report Results. Source: Guide Section 2.1.3
4. Correct answer: A. Defining the future state includes ensuring it is clearly defined, achievable with available resources, that key stakeholders have a shared vision developed by consensus, and that measurable objectives are established to ensure the desired vision is met. Why others wrong: B - fixing regression coefficients is detailed analysis work, not future state definition; C - the Guide requires a shared vision by consensus, not a single person deciding alone; D - cleansing and joining data is data preparation in Analyze Data. Source: Guide Section 2.1.4
5. Correct answer: A. In the Guide example for improving customer experience, the question Will customer experience improve by adding a new feature such as a pay wallet is mapped to prescriptive analytics, because research questions are tied to analytics types. Why others wrong: B - descriptive maps to the questions about what the factors are and how to measure experience; C - the Guide example does not present a diagnostic why question here; D - predictive maps to classifying an individual transaction as positive or negative, not to evaluating a proposed feature. Source: Guide Section 2.1.5
6. Correct answer: A. When complexity is more than anticipated, initiatives are typically implemented in multiple stages of maturity: a proof of concept focused on feasibility, a pilot focused on a limited scale solution to discover integration and quality issues, and a production stage focused on business value. Why others wrong: B - the Guide sequences feasibility and pilot before production; C - planning is an iterative process and the approach changes as new knowledge is gained; D - the Guide does not endorse skipping proof of concept and describes the staged sequence as typical. Source: Guide Section 2.1.6
7. Correct answer: A. The Guide lists Prioritization among techniques for Identify the Research Question, used throughout the effort to focus attention on the most urgent items, for example when formulating research questions, sharing insights, and recommending actions. Why others wrong: B - data preprocessing and normalisation is an Analyze Data preparation step; C - boxplot reading is an exploratory or reporting visualisation technique; D - functional decomposition is used to develop an implementation plan, not to formulate research questions. Source: Guide Section 2.1.7
8. Correct answer: A. Structured data is organized and formatted, such as data in a database management system accessed by a query language like SQL, while unstructured data exists outside any organized repository and takes forms such as text from documents, emails, and social media. Why others wrong: B - it reverses the two definitions; C - free-text emails and social posts are unstructured even though stored digitally; D - SQL warehouse records are structured by the Guide definition. Source: Guide Section 2.2.1
9. Correct answer: A. The Guide defines Veracity as referring to the trustworthiness of the data and that which presents uncertainties and inconsistencies in the data, which is exactly what the questionable sensor feed shows. Why others wrong: B - Value refers to driving analytics from real, valuable business goals, not trustworthiness; C - volume concerns quantity, which is not the issue described; D - velocity concerns speed of arrival, not trustworthiness. Source: Guide Section 2.2.2
10. Correct answer: A. Passive data collection is unobtrusive collection from users in their day-to-day transactions, such as point-of-sale data, while active data collection actively seeks information from stakeholders for a specific goal, such as surveys and self-reports using rating scales like the Likert scale. Why others wrong: B - it reverses the definitions; C - the survey is active, not passive; D - point-of-sale logs already exist from daily transactions, making them passive. Source: Guide Section 2.2.3
11. Correct answer: A. The Guide defines Completeness as the data being comprehensive and including what is expected with nothing missing, and says completeness might be assessed by ensuring required fields do not include null values. Why others wrong: B - Accuracy is about data being correct and not misleading, assessed by comparing a front-end system with the database; C - Consistency concerns reliability across sources, not null fields; D - Veracity is a Source Data characteristic about trustworthiness, not the named validation characteristic for missing required fields. Source: Guide Section 2.2.4
12. Correct answer: A. The Guide states that when validating data analysts use data mapping to create a source-to-target data map that defines the mapping between the data sources being used and the target system. Why others wrong: B - business rules analysis provides understanding of the rules governing the data, guiding what should be validated, not the field correspondence itself; C - functional decomposition is an implementation-planning technique; D - root cause analysis is an Interpret and Report Results technique, not a Source Data mapping technique. Source: Guide Section 2.2.5
13. Correct answer: A. The Guide says the business analysis professional provides insights into the plan or may draft the initial plan, but it is the data scientist who possesses deep technical expertise to decide how the data analysis will be conducted. Why others wrong: B - the analyst may draft the plan but does not make the final technical decision; C - the sponsor funds the work but does not decide the analytical technique; D - the reporting team communicates results, it does not choose the analysis method. Source: Guide Section 2.3.1
14. Correct answer: A. Preparing data includes understanding the relationships that exist between data, for example whether two tables have a 0 to 1, 1 to 1, or 1 to many relationship, and establishing the joins or linkages between sources to create a coherent dataset. Why others wrong: B - trustworthiness with uncertainties is veracity from Source Data, not relationship cardinality; C - communication needs belong to Interpret and Report Results; D - ranking solution options belongs to Use Results to Influence Business Decision Making. Source: Guide Section 2.3.2
15. Correct answer: A. Exploring data involves an initial exploratory analysis to ensure the data collected is what was expected from the sources, providing a form of quality check to ensure the right type and quality of data is being obtained prior to executing more detailed analysis. Why others wrong: B - producing the final report is in Interpret and Report Results; C - approving sources and acceptance criteria is data validation in Source Data; D - ranking solution options is in Use Results to Influence Business Decision Making. Source: Guide Section 2.3.3
16. Correct answer: A. The Guide states that when variables are compared against prices, most predictors had a linear relationship with price, where some change in the predictor variable changes the price by a proportional amount in a straight-line relationship. Why others wrong: B - a categorical relationship with unordered labels does not match the proportional straight-line description; C - no association contradicts the observed proportional movement; D - bimodality describes two peaks in a distribution, not a proportional predictor-to-price relationship. Source: Guide Section 2.3.4
17. Correct answer: A. Assessing the analytics approach is performed iteratively, and even when exploration is acceptable the analysis may fail to answer the questions; in these scenarios the data exploration and data analysis tasks are repeated, iterating until the data scientist is comfortable with the data sources and their value toward answering the research questions. Why others wrong: B - proceeding to recommendations when the question is unanswered ignores the required iteration; C - the Guide describes iterating and adapting, not permanently stopping; D - presenting an unanswered question as a conclusion contradicts letting the data answer the question. Source: Guide Section 2.3.5
18. Correct answer: A. Stakeholder analysis is ongoing and continually updated, so the team re-validates changing needs such as how quickly analytics results are expected; the decrease is 20 minus 15 equals 5, and 5 divided by 20 equals 0.25, a 25 percent decrease. Why others wrong: B - 5 is the absolute drop in days, not the percentage; C - 33 percent divides the 5-day drop by the new value 15 instead of the original 20; D - 75 percent is the remaining proportion 15 over 20, not the decrease. Source: Guide Section 2.4.1
19. Correct answer: A. When planning stakeholder communication the Guide says communications can involve an intermediate or final result since analytics initiatives are inherently iterative, and that stakeholders should be kept informed about progress throughout. Why others wrong: B - the Guide says formality and level of detail may vary amongst stakeholders, not be identical; C - the Guide says to consider the level of privacy and confidentiality to be maintained; D - the Guide says to record responses and feedback for further action and follow-up. Source: Guide Section 2.4.2
20. Correct answer: A. Determining communication needs captures stakeholder preferences such as what information is most relevant to them, how they wish to receive information, how often they wish to be updated, who the decision-makers are, and what biases they carry. Why others wrong: B - hyperparameters concern Analyze Data, not communication preferences; C - source-to-target data mapping is a Source Data validation technique; D - the five data quality characteristics belong to data validation in Source Data. Source: Guide Section 2.4.3
21. Correct answer: A. In the Guide example, Uber ride density correlated with crime rate, but although it is a fascinating correlation, demand prediction should not be modelled on the crime rate without stronger evidence of a relationship; analysts use sound statistical judgment to translate patterns into true insights. Why others wrong: B - the Guide explicitly warns against treating such correlation as causal; C - the Guide derives a useful observation from the merge rather than discarding it; D - the Guide requires appropriate analysis to confirm business relevance before calling a pattern an insight. Source: Guide Section 2.4.4
22. Correct answer: A. The Guide visualization best practices state that when there is only a single or a couple of metrics involved, simple text may be a more effective way to communicate the metrics, for example ROI, profits, percentage, and average values. Why others wrong: B - complicated graphs used for visual appeal only may end up complicating the message; C - the Guide says a pie chart where the audience must interpret arc lengths and angles can be replaced with a simpler form; D - extra dimensionality for appeal complicates rather than clarifies a single metric. Source: Guide Section 2.4.5
23. Correct answer: A. The Guide describes The Big Idea as a technique used to communicate the most relevant findings in a succinct manner, giving the example of an S-curve between marketing expense and revenue where the big idea answers the so-what question that beyond a certain spend marketing has no impact on revenue. Why others wrong: B - data mapping is a Source Data technique; C - functional decomposition supports implementation planning; D - data profiling is a Source Data quality technique, not a reporting message technique. Source: Guide Section 2.4.6
24. Correct answer: A. In the Guide case study response, the data science team found Facebook had the highest regional adoption, which was conducive to the research problem, so they selected Facebook as the optimal medium. Why others wrong: B - the case concerns social media measurement, not a printed flyer; C - the team did not select email newsletters; D - the team selected Facebook based on highest regional adoption, not radio. Source: Guide Section 2.4.7
25. Correct answer: A. The Guide says when analysis provides valuable insights the effort switches to using results to drive change, and these possibilities are referred to as solution options, which may include elements of process, tool, resource, or IT system changes that analysts rate, rank, and propose as a recommendation. Why others wrong: B - research questions are formulated in Identify the Research Question before analysis; C - data quality characteristics belong to data validation in Source Data; D - communication preferences belong to Interpret and Report Results. Source: Guide Section 2.5.1
26. Correct answer: A. When developing an implementation plan analysts break down the work to implement the proposed changes, and functional decomposition is the technique used to drill high-level tasks down into lower-level tasks and activities, often as a work breakdown structure or story map. Why others wrong: B - Likert-scale survey design is an active data collection technique in Source Data; C - data cleansing is a data preparation step in Analyze Data; D - boxplot interpretation is a visualisation technique, not implementation planning. Source: Guide Section 2.5.2
27. Correct answer: A. The Guide says business analysts are well suited to fulfil the change manager role because they are able to ensure the continuity between the analytics work and implementation, overseeing the transformation of analysis results into implemented policies and procedures. Why others wrong: B - deep technical model-building is the data scientist's expertise, not the basis for the change manager role; C - the Guide says stakeholders agree on what changes to make before implementing, not that the analyst overrides them; D - setting data quality acceptance criteria is a Source Data validation activity. Source: Guide Section 2.5.3
28. Correct answer: A. The Guide lists Benchmarking and Market Analysis as a technique used to evaluate competitor and market reactions if certain recommendations from the analytics initiative are implemented. Why others wrong: B - data profiling is a Source Data quality technique; C - exploratory data analysis belongs to Analyze Data; D - data mapping is a Source Data validation technique, not for evaluating market reactions to recommendations. Source: Guide Section 2.5.4
29. Correct answer: A. The Guide case study states that to differentiate in a highly competitive environment, several top retailers are investing in more customized gift offerings that are personalized for newer target market segments. Why others wrong: B - the case describes the online gifting industry, not withdrawal from online channels; C - the case emphasises customization and personalization, not uniform price cuts; D - the case describes increasing effective marketing practices, not halting marketing. Source: Guide Section 2.5.5
30. Correct answer: A. The Guide says before recommending changes an evaluation determines the success of the analysis, asking whether the outcome answered the research question and how well it addressed the business need; when the data does not deliver required insights and no feasible solution is found, the analytics cycle is repeated starting with a new research question. Why others wrong: B - dashboard colour is a reporting detail, not the evaluation the Guide describes; C - the Guide does not publish a recommendation when no feasible solution exists; D - liking the visuals is not the evaluation criterion for recommending actions. Source: Guide Section 2.5.1
31. Correct answer: A. The Guide says developing an implementation plan includes a sequence showing flow and task dependencies, and that constraints, assumptions, risks, and dependencies are also identified and discussed, with the work broken down using functional decomposition. Why others wrong: B - veracity is a Source Data characteristic, not an implementation-plan element; C - Likert-scale survey wording is an active data collection detail; D - choosing the statistical test is an Analyze Data decision made by the data scientist. Source: Guide Section 2.5.2
32. Correct answer: A. The Guide describes the centralized model as the analytics team operating as a single unit supporting other business units in decision-making, with an analytics Centre of Excellence as a good example. Why others wrong: B - the decentralized model embeds analytics into different business units, not a single unit; C - the hybrid model is a mix of centralized and decentralized such as a hub and spoke, not solely embedded teams; D - proof of concept is a maturity stage, not a standing organizational model. Source: Guide Section 2.6.1
33. Correct answer: A. The Guide says three major components form the pillars of a robust talent strategy: establishing the right team structure for analytics initiatives, the ability to create an eco-system for learning and development, and establishing best practices for analytics initiatives. Why others wrong: B - selecting a predictive model is an Analyze Data technical decision, not a talent-strategy pillar; C - approving data sources and acceptance criteria is Source Data validation; D - data mapping is a Source Data technique, not a talent-strategy pillar. Source: Guide Section 2.6.2
34. Correct answer: A. The Guide lists data governance as a planning consideration in an organization-level data strategy, defining it as the rules and policies that manage the data assets of an organization to ensure high-quality data. Why others wrong: B - data architecture is the models and standards governing how data is collected, stored, and integrated, not the governing rules and policies; C - data security protects data for privacy and confidentiality; D - metadata management administers information maintained about the data assets, not the governing rules. Source: Guide Section 2.6.3
35. Correct answer: A. The Guide lists Balanced Scorecard among techniques for the organizational-level strategy domain, used to describe a balanced view of the organization from different perspectives and useful for aligning the data strategy to business objectives and outcomes. Why others wrong: B - a boxplot is an exploratory and reporting visualisation, not a strategy-alignment technique; C - data cleansing is an Analyze Data preparation step; D - a Likert-scale survey is an active data collection technique in Source Data. Source: Guide Section 2.6.4
How did you score?
30-35: you are in strong shape - book the exam. 25-29: passing range, tighten up your weakest domain. Below 25: revisit the Guide to Business Data Analytics chapters for the domains you missed most, then retest. For more practice across the full CBDA blueprint, explore the rest of our CBDA materials on certificate.tips.
