IIBA CBDA - Questions & Answers

Predictive analytics is a type of analytics that uses historical and current data, as well as statisticaland machine learning techniques, to forecast future events or outcomes, such as product purchases, customer behavior, or market trends 12. To answer the question `How many product purchases are expected this
coming year given the current marketing campaign?', predictive analytics would be performed to estimate the demand and sales based on the existing data and the marketing campaign variables.

A well-defined objective is one that is specific, measurable, achievable, relevant, and time-bound (SMART) 1. The goal of increasing sales of a mature product by reducing its price by 20% which would result in increased revenues of 2% over a 6-month period meets all these criteria, as it clearly states what the desired outcome is, how it will be measured, whether it is realistic and attainable, how it aligns with the organization's strategy, and when it will be achieved

According to the Business Data Analytics (IIBA - CBDA) principles, extreme outliers in a dataset can be identified visually on a scatter plot as points that are distinctly separate from the bulk of the data. In this visual, there are three points that are significantly higher on the y-axis (credit card expense) relative to their position on the x-axis (household income), indicating unusual credit card purchases.
References:
The identification and interpretation of outliers is a standard practice in data analytics and is covered under the Business Data Analytics (IIBA - CBDA) learning resources.

The analytics team should restart the work with formation of a new research question, because the existing one may not be well-defined, relevant, or feasible. A well-formed research question is the first step of the business data analytics cycle, and it guides the subsequent steps of sourcing, analyzing, interpreting, and reporting data. If the data analysis does not yield meaningful insights, the team should revisit the research question and refine it based on the business problem, stakeholder needs, data availability, and analytical methods.

A research question is a clear and focused question that guides the data analytics process and defines the expected outcome or value of the analysis 1. A research question that is too broad may lead to the concern of being difficult to identify the key performance indicators (KPIs) to measure, as KPIs are specific, quantifiable, and relevant metrics that indicate the progress and success of the analysis in relation to the research question 23. A broad research question may also result in too much or too little data, unclear or conflicting objectives, or irrelevant or ambiguous results

A justifiable concern at this stage of the team's efforts is changing the mindsets of business stakeholders to trust insights gleaned from data over experience and intuition. This is because some stakeholders may have a strong attachment to their own opinions or beliefs, and may resist or ignore data that contradicts them. This can create a barrier to data-driven decision making, which requires a culture of curiosity, openness, and evidence-based reasoning. The team needs to communicate the value and validity of their data analysis, and persuade the stakeholders to adopt a data-driven mindset

A funnel chart is a type of chart that shows the values of different stages of a process, such as a sales pipeline, where each stage represents a subset of the previous one. A funnel chart is useful for showing the conversion rate, the drop-off rate, and the potential revenue or profit at each stage. A funnel chart would be an effective way to present the data requested by the call center, as it would show the number of leads, calls, customers, and sales, as well as the percentage of change between each stage.

Providing `one source of the truth' means ensuring that there is a single, consistent, and authoritative source of data that can be used for analytics and decision making across the organization. This objective can be achieved by enforcing master data management principles and practices, which involve defining, governing, and maintaining the quality and integrity of the core data entities that are shared by multiple systems and processes. Master data management helps to eliminate data silos, reduce data duplication and inconsistency, and improve data accuracy and reliability

A pie chart is a circular chart that shows the proportion of each category in a whole by dividing the circle into slices. A pie chart would result in confusion when comparing the number of sales for a specific model of electric car across all 50 US states, because it is difficult to compare the angles and areas of the slices, especially when there are many categories with similar values. A pie chart also does not show the absolute values of each category, unless they are labeled or annotated. A better alternative would be a bar chart, which can show the number of sales for each state along a common axis, making it easier to compare and rank the values