BSAN2201 Chap.1 Business Analytics Decisions
Business Analytics Decisions
Business analytics is the connection between business data, analysis and a decision. This chapter begins with the decision owner and the action that could change, then separates business intelligence, data science and data analytics by their dominant purpose.
It builds the three-part analytics ladder: descriptive work identifies what happened, predictive work estimates what may happen, and prescriptive work recommends what to do next. The central skill is not memorising labels. It is matching the required output to the business question and making the path from evidence to action explicit.
A useful analytical question names the actor, choice, outcome and time horizon.
It should also create a real fork: if a different result would not change the action, the proposed analysis may be background rather than the decision problem. The chapter therefore treats descriptive, predictive and prescriptive work as connected but distinct outputs. A pattern can motivate a forecast, and a forecast can feed an optimisation, yet each stage has its own assumptions and stopping point.
The closing drills require students to identify what the evidence directly supports, which decision it informs and what additional observation would reverse the recommendation.
What this chapter covers
- 01
Decision owners and business questions
- 02
Business intelligence and business analytics
- 03
Descriptive analytical purpose
- 04
Predictive analytical purpose
- 05
Prescriptive analytical purpose
- 06
Evidence-to-action boundaries
Worked example · free
Classify a staffing decision
- +1Summarising late orders by time and zone is descriptive.
- +1Estimating Friday demand is predictive.
- +1Choosing drivers under the wage limit is prescriptive.
- +1Explain that the forecast becomes an input to the constrained allocation.
Key terms
- Decision owner
- The person or role accountable for acting on an analytical result.
- Business intelligence
- Reporting and summarisation that show what has happened or is currently happening.
- Descriptive analytics
- Analysis that identifies patterns in current or historical data.
- Predictive analytics
- Analysis that estimates a future value, event or category.
- Prescriptive analytics
- Analysis that recommends an action under an objective and constraints.
Business Analytics Decisions FAQ
How do I identify the analytical type in a scenario?
Look at the required output. A summary or pattern is descriptive, an estimate of a future outcome is predictive, and a selected action is prescriptive. If the scenario contains several stages, label each stage and explain which output directly answers the manager's decision.
Can one business problem use all three types?
Yes. A team may describe historical demand, predict next week's volume and then optimise staffing or inventory. The types form a dependency chain, but they should not be blended into one vague label because each stage produces a different output and has different assumptions.
Why is a dashboard not automatically business analytics?
A dashboard can provide useful descriptive evidence, but the analytics chain is incomplete until the evidence is interpreted for a business choice. State who uses the output, what action can change and which measure or threshold would alter that action.
How can I tell whether a question is decision-ready?
A decision-ready question names the person or role choosing, the alternatives within that person's control, the outcome that matters and the time horizon. It also identifies what evidence could change the choice. If a question asks only what happened, add the action or threshold that makes the description consequential.
Assessment move
For each lecture scenario, write a one-sentence decision statement before naming a method. Then classify the output as a pattern, forecast or action. Practise mixed scenarios where a forecast feeds an optimisation, because those make the boundary visible. Finish every answer with a sentence connecting the output to the decision and a limitation that could reverse it.
Build a four-column retrieval sheet headed question, output, analytical type and business action. After answering, alter one fact and decide which column changes first. Reverse the expected result and ask whether the manager's choice would also reverse. If it would not, refine the decision statement or identify the threshold that actually matters.
Once a week, take a dashboard example and write the missing predictive or prescriptive step without pretending that the descriptive evidence proves a cause. This routine trains boundaries and prepares the language needed for the later case-study judgement. Add a fifth column for the population and period represented by the evidence.
Then write a deliberately weak conclusion and repair it by naming the missing comparator, uncertainty or implementation condition. Shuffle the scenarios and classify them without headings so recall depends on the decision logic rather than slide order. End the session by teaching one distinction aloud, using a fresh example: descriptive versus predictive, predictive versus prescriptive, or evidence versus action.
If the example cannot survive a changed fact, revisit the decision statement and threshold. Keep one error log that records the mistaken category, the clue that should have changed the answer and a corrected example. Review that log before attempting a new mixed scenario so recurring boundary errors become visible.
Working through Business Analytics Decisions in BSAN2201? Sia is AskSia’s AI Business and Economics tutor — ask any BSAN2201 Business Analytics Decisions question and get a clear, step-by-step explanation grounded in how BSAN2201 is taught and assessed. Read this chapter free, then take your hardest questions to Sia.