SingaporeBusiness ManagementSyllabus dot point

What do managers actually do, and how should they make decisions under uncertainty?

Explain the functions and roles of management and analyse approaches to decision making, including scientific versus intuitive methods and the use of decision trees

A focused answer to the H2 Management of Business outcome on management and decision making. The functions and roles of management, scientific versus intuitive decision making, the decision-making process, and the use and limits of decision trees with a worked expected-value calculation.

Generated by Claude Opus 4.89 min answerUpdated 2026-06-06

Reviewed by: AI editorial process; not yet individually human-reviewed

Have a quick question? Jump to the Q&A page

Jump to a section

What this dot point is asking
The answer
Examples in context
Try this

What this dot point is asking

SEAB wants you to explain what managers do and to analyse how they make decisions, including the contrast between scientific (data-driven) and intuitive (judgement-based) approaches and the use of quantitative aids such as decision trees. The exam rewards being able to calculate an expected value and then critically judge how much weight the result deserves.

The answer

The functions of management

A long-standing summary of what managers do is planning, organising, commanding (directing), coordinating and controlling - setting objectives and plans, arranging resources and people, directing and motivating staff, coordinating activities, and monitoring results against targets. A complementary view describes managerial roles - interpersonal (figurehead, leader, liaison), informational (monitor, disseminator, spokesperson), and decisional (entrepreneur, disturbance handler, resource allocator, negotiator). Both stress that management is about achieving objectives efficiently through other people.

Scientific versus intuitive decision making

Scientific (rational) decision making follows a structured process: define the objective, gather and analyse data, generate and evaluate options against criteria, choose, implement, then review. It uses tools like forecasts, investment appraisal and decision trees. It is rigorous and defensible but can be slow, data-hungry, and falsely precise if the data is poor.
Intuitive decision making relies on the manager's experience, judgement and instinct. It is fast and can capture tacit knowledge that data misses, but it is subjective, hard to justify, and prone to bias.

Good managers blend the two: use analysis where data and time allow, and judgement where they do not.

The decision-making process

A typical structured process runs: identify the problem or objective; gather information; develop alternatives; evaluate them against criteria (cost, risk, fit with objectives); select; implement; and review the outcome to learn. Skipping the review step is a common organisational weakness.

Decision trees

A decision tree is a diagram that maps decisions (squares) and chance events (circles) with their probabilities and payoffs, used to compute the expected value of each option:

EV = \sum (\text{probability} \times \text{payoff})

The firm chooses the option with the higher expected value, net of any cost. Trees force explicit estimates and structured comparison - their strength - but they are only as reliable as the estimated probabilities and payoffs, they reduce a decision to an average that ignores attitude to risk, and they exclude qualitative factors.

Evaluating decision aids

Quantitative tools should inform, not replace, judgement. A positive expected value is a guide, but the manager must still weigh data quality, the firm's ability to absorb the downside, and the qualitative factors the model omits. The exam rewards calculating correctly and then critiquing the number's limits.

Worked example

A firm chooses between two marketing campaigns. Campaign A: 70% chance of \ $500{,}000 profit, 30% chance of \$ 100{,}000 profit, costing \ $80{,}000. Campaign B: 50% chance of \$ 900{,}000 profit, 50% chance of \ $0, costing \$ 120{,}000. Use expected values to recommend, then evaluate.

Step 1: Expected value of Campaign A outcomes

EV_A = (0.7 \times 500{,}000) + (0.3 \times 100{,}000) = 350{,}000 + 30{,}000 = \$380{,}000

Net of cost: $380{,}000 - 80{,}000 = \$ 300{,}000$.

Step 2: Expected value of Campaign B outcomes

EV_B = (0.5 \times 900{,}000) + (0.5 \times 0) = 450{,}000 + 0 = \$450{,}000

Net of cost: $450{,}000 - 120{,}000 = \$ 330{,}000$.

Step 3: Compare on expected value

Campaign B has the higher net expected value (\ $330,000 versus \$ 300,000), so on the numbers alone B is preferred.

Step 4: Evaluate beyond the number

The margin is small (\ $30,000) and B is far riskier - a 50% chance of \$ 0 return versus A's guaranteed-positive outcomes. A cash-constrained or risk-averse firm might rationally prefer the safer Campaign A despite its slightly lower EV, because the downside of B (no return on $120,000 spend) could hurt. The recommendation therefore depends on the firm's risk appetite and the reliability of the probabilities, illustrating that EV guides but does not settle the decision.

Examples in context

Example 1. Data-driven decisions at scale. Large e-commerce and ride-hailing platforms make pricing, inventory and routing decisions using vast data and algorithms - a highly scientific approach where data is abundant and decisions are repeated millions of times. Yet senior leaders still rely on judgement for one-off strategic moves (entering a new country, a major acquisition) where data is thin and the future uncertain, showing the rational-intuitive blend across different decision types.

Example 2. A retailer's new-store decision. A Singapore retail chain deciding whether to open in a new mall might build a decision tree weighing footfall forecasts, rent and competitor presence to compute an expected return. But the final call also rests on judgement about the mall's trajectory, brand fit and management capacity - factors the tree cannot capture. The store decision illustrates using a quantitative aid to structure the choice while reserving judgement for what the numbers omit.

Try this

Q1. State three functions of management. [3 marks]

Cue. Any three of: planning (setting objectives and plans), organising (arranging resources and people), directing or commanding (leading and motivating staff), coordinating (aligning activities), and controlling (monitoring results against targets).

Q2. A project has a 0.4 chance of a \ $200{,}000 gain and a 0.6 chance of a \$ 50{,}000 loss. Calculate its expected value. [3 marks]

Cue. $EV = (0.4 \times 200{,}000) + (0.6 \times -50{,}000) = 80{,}000 - 30{,}000 = \$ 50{,}000$. The positive expected value suggests proceeding, subject to risk and data quality.

Q3. Analyse why a manager might reject the option with the highest expected value. [6 marks]

Cue. Expected value is an average that ignores the spread of outcomes and the firm's attitude to risk, so a high-EV option may carry a large probability of a damaging loss that a cash-constrained or risk-averse firm cannot absorb. The probabilities and payoffs may also be unreliable estimates, and qualitative factors (brand fit, reputation, staff capacity) may favour a different option. A manager may therefore rationally choose a lower-EV but safer or better-fitting option, treating EV as one input among several.

Exam-style practice questions

Practice questions written in the style of SEAB exam questions on this dot point, with worked answer explainers. The year tag is the paper they imitate, not the source.

Original8 marksA retailer must decide whether to launch a new product. A market research report estimates the chance of high demand at

0.6

(profit \

800{,}000) and low demand at

0.4

(loss \

300{,}000); launching costs \$100{,}000 to develop. Using a decision tree, calculate the expected value of launching and discuss whether the firm should proceed.

Show worked answer →

Calculate the expected value of the launch outcomes:

EV = (0.6 \times 800{,}000) + (0.4 \times -300{,}000) = 480{,}000 - 120{,}000 = \$360{,}000

Net of the \ $100,000 development cost, the net expected value is$ 360{,}000 - 100{,}000 = \ $260{,}000$ . Since this is positive, the quantitative decision-tree result favours launching.

Discuss the limitations before deciding. The probabilities and payoffs are estimates from a forecast that may be wrong, so the EV is only as good as its inputs. EV is an average that ignores the firm's attitude to risk - a 40% chance of a $300,000 loss could threaten a firm with weak cash reserves even though the average is positive. The model also ignores qualitative factors (brand fit, competitor response, staff capacity).

Reach a judgement. On the numbers the firm should launch, since the expected value is clearly positive. But the recommendation should be conditioned on the reliability of the estimates and the firm's ability to absorb the downside; a cash-strong firm should proceed, while a fragile one might hesitate or reduce the risk (a smaller launch, more research). A strong answer uses the EV as a guide, not a verdict, and weighs risk and data quality.

Markers reward a correct EV calculation net of cost, recognition that EV is an average ignoring risk attitude and data quality, and a judgement that treats the tree as a decision aid conditioned on context.

Original6 marksExplain the difference between scientific and intuitive decision making, and analyse one situation in which intuition might be preferable.

Show worked answer →

Explain the distinction. Scientific (or rational) decision making follows a structured process - setting objectives, gathering and analysing data, evaluating options against criteria, and choosing systematically (using tools like decision trees and forecasts). Intuitive decision making relies on the manager's experience, judgement and gut feeling rather than formal analysis.

Analyse when intuition may be preferable. Intuition can be better when a decision must be made very quickly with no time to gather data (a crisis), when reliable data simply does not exist (a genuinely novel situation), or when the experienced manager's tacit knowledge captures subtleties that data misses. In such cases insisting on full scientific analysis would be too slow or impossible, and seasoned judgement outperforms incomplete numbers.

Markers reward a clear contrast between structured data-driven and experience-based judgement, and a developed situation (speed, absent data, or tacit expertise) where intuition is justified.