Decision trees — Edexcel A-Level Business
Verified courseTeaching note checked against the official specification.
Decision trees explained
Decision trees covers 3 specification points, including A) Construct and interpret simple decision… and B) Calculations and interpretations of figures….
Read the full explanation
Open a point below for its explanation and revision guidance.
a) Construct and interpret simple decision tree diagrams
The diagram runs left to right from a square decision node, where each branch is a course of action the business could take, into circular chance nodes whose branches carry probabilities that must total one. The money an outcome is worth sits at the right hand tip, the cost of taking an option sits on its own branch, and the whole thing is then valued back from right to left. Drawing it forces managers to state the options, the possible states of the world and the odds attached to each, which is the real purpose: a new product launch at a firm like Greggs becomes a comparison of two payoffs rather than an argument between departments. The shape is also where evaluation begins, because a branch nobody drew can never be chosen.
b) Calculations and interpretations of figures generated by these techniques
Value the tree from right to left. At each chance node multiply every payoff by its probability and add the products, giving an expected value in pounds, then subtract the cost written on the branch running into that node to reach the net gain; at the decision node take the branch with the highest net gain and cross the others off. A net gain of 180,000 pounds against 140,000 pounds says the first option is worth roughly 40,000 pounds more on a weighted average of the possible futures, not that the firm will bank that sum, because a single launch delivers one branch only. That gap between a weighted average and a real outcome is where the interpretation marks sit, and a narrow gap should make any recommendation tentative.
c) Limitations of using decision trees
The arithmetic is exact, and that is the danger: the method turns two guesses, the probability and the payoff, into a single confident looking figure. Probabilities usually come from past data or a manager's judgement, and the manager building the diagram also decides which branches exist, so a preferred option can be flattered by omitting an unattractive outcome or shading the odds. Expected value is risk neutral, so it says nothing about whether a small firm could survive the worst branch, nor about staff reaction, reputation or ethics. Outcomes are treated as a short list of discrete states when demand is really a continuous range, and the estimates date quickly in a volatile market, which is why the technique suits repeated routine choices better than one off strategic ones.
Your focus
- a) Construct and interpret simple decision tree diagrams
- Calculate expected values and net gains from a provided decision tree diagram.
- Interpret the calculated net gains to recommend the most financially viable option.
Show all 5 objectives
- Evaluate the limitations of decision tree calculations, recognising that expected values are probability-weighted averages rather than guaranteed outcomes.
- c) Limitations of using decision trees
Decision trees exam tips
Marking Points
- Using a square for a decision node and a circle for a chance node, with every branch labelled and the probabilities at each chance node summing to one.
- Putting the cost of an option on its branch and the payoff at the end of each outcome branch, in the money units the case study uses.
- Reading the diagram back correctly, stating which option the business would take and why the alternative was rejected.
- Explaining that laying the options and risks out explicitly makes the choice defensible to shareholders and easier to revisit later.
- Show the expected value working, probability multiplied by payoff and summed across every branch of the chance node, with money units kept throughout.
- Deduct the cost of the option to reach net gain before comparing options, rather than comparing raw expected values.
- Interpret the winning figure as a probability-weighted average that assumes the estimates are right, not as a guaranteed return.
- Comment on how close the competing net gains are, since a narrow margin makes the decision sensitive to a small change in the probabilities.
- Attacking a named input, such as the probability attached to a strong market response, and saying who produced it and why it may be optimistic.
- Pointing out that expected value ignores the scale of the downside, so a business with weak liquidity may rationally reject the branch with the highest figure.
- Noting that whoever draws the diagram selects the options and can build in bias, deliberately or not.
- Balancing the criticism with what the technique still does well, which is making assumptions visible, comparable and open to challenge.
Examiner Tips
- 💡Diagram questions are low tariff and marked on mechanics, so label every branch, probability and payoff even when you are short of time.
- 💡Interpretation earns more than drawing, so finish with a sentence naming the option chosen and the figure behind it.
- 💡If the stem gives probabilities as percentages, convert them to decimals on the diagram and keep one convention throughout.
- 💡Set the multiplication for each branch out on its own line to clearly demonstrate the method used.
- 💡Follow through applies, so an early arithmetic slip can still support a valid interpretation provided you argue consistently from your own figure.
- 💡Expect to be asked whether the business should proceed, which needs the calculated number plus at least one qualitative factor the tree cannot price.
- 💡This is evaluation territory, so pair each limitation with a condition, such as the estimates being sound where the firm has years of comparable sales data.
- 💡Use the figures you calculated earlier in the question to show how a change in one probability would reverse the decision.
- 💡A strong conclusion states which limitation matters most to this particular firm and justifies that ranking.
Common Mistakes
- Probabilities at a chance node that do not add to one, usually because an outcome such as no change in demand has been left off the diagram.
- Muddling the two node shapes, so the diagram shows the business choosing an event it cannot control, such as the state of the economy.
- Netting the cost off the payoff at the tip and again on the branch, which counts the same outlay twice.
- Adding probabilities to payoffs instead of multiplying them, which yields a figure very close to the original payoff rather than the correct expected value.
- Forgetting to subtract the cost of the option, so an expensive branch with a big payoff wins on paper.
- Rounding part way through and carrying the rounded figure onward, which can shift the final comparison between two close options.
- Saying it is unreliable because the future is uncertain, which is true of every forecasting tool and earns no development.
- Claiming the technique ignores cost when the cost of each option is written on its branch, the real omissions being qualitative factors and risk appetite.
- Rejecting the method outright instead of judging when it helps, for example for a repeated low value decision where averages genuinely apply.