About Decision Trees

What is a decision tree?

A decision tree is one approach to decision making which can be a useful tool for certain types of decisions, especially where the financial costs and pay offs are crucial determining factors.

It’s called a decision tree because drawing lines from the initial choice out to different options, each of which may involve further choices and options produces a tree like branched diagram. These ‘trees’ are often lying on their side as the diagrams are usually drawn from left to right.

When decision trees are most useful

When you have a decision to make between several options where the costs and the potential financial returns can be estimated reasonably realistically.

A decision tree can be a great way to layout out the options in a visual format that everyone can grasp. This makes it a particularly useful tool when several people are involved in making the decision, or at least need to have an understanding of how it was made.

Business decisions on which product or service to prioritise to boost profits, which new development is likely to be most profitable, whether a contract is worth pitching for – decision trees can be a good tool to bring clarity to decisions of this kind. They can sometimes be useful for personal decisions too, when the choice hinges on the financial costs and benefits of several different options to choose from, e.g. which car to buy, which flat to rent.

Creating a decision tree forces you to attempt to get real about the financial implications, even though most of the figures you use are likely to be best guesses.

How to create a decision tree

Squares, circles and triangles are standard symbols in drawing decision trees, linked by lines. To create a decision tree begin with the initial decision. All decisions are drawn as a square. If you’re doing this by hand use a large board or sheet of paper and center the square representing the main core decision horizontally on the left side.

Then draw lines fanning out from the square to represent the main potential options. Remember, doing nothing can sometimes be a valid option. Different options can lead to further decisions (shown as squares) or to ‘events’ or outcomes that are not certain, shown as circles.

From secondary decision squares draw more lines to represent the options to choose from. From circles representing uncertain outcomes draw lines that represent possible outcomes. Lines that end in a projected definite outcome are represented by a triangle.

Write above each line the option or the outcome it represents. Then flesh out your tree with financial cost and pay off figures. Some figures may be known; otherwise you will have to estimate them.

To work out the best potential choice requires an estimation of probability – how likely do you believe each possible outcome will be. This probability weighting is crucial for the best possible decision your decision tree will indicate. Allocate each outcome radiating out from a circle with a value that is either a % or a decimal point. Probability guesstimates for each branch from a circle must add up to either 1 or 100 in total.

Now to calculate the projected financial implications of all the possible choices you have when making your decision. Take the end anticipated pay off x its probability – the cost of implementing that option.

Working backwards from right to left, calculate the total figure for all the outcomes for each option you have included. A comparison of these figures will reveal your best choice from the options.

Example of a decision tree

Since decision trees are in essence a visual financial calculating tool, it’s much easier to show how they work with a simple illustration.


Advantages and limitations of decision trees

Often just the process of creating your tree diagram prompts you to include options you may not otherwise have thought of. And then it focuses you on evaluating all the options.

Decision trees can also demonstrate in an easily recognisable way the financial advantages and disadvantages of the different choices that can be made – useful if you need to justify the decision you take to others, or if you want a quick way to show the people you need to implement the decision why it makes financial sense.

Although they can be drawn to look impressively professional and convincing, be prepared to explain and back up the figures you’ve used for costs, payoffs and probabilities. Like computer programmes, a decision tree is only as good as the information you put into it!

Frequently the figures used​ can only be guesstimates. The estimates of probability are just that – the best guesses we can make between future outcomes using the information available with our knowledge and experience.

Unfortunately we often have a tendency to conflate figures written down with verifiable facts! Since you can’t predict the future with complete accuracy, it’s important not to get carried away with your calculations. Remember they are based on your assumptions and hopes; otherwise a harsh blast of reality may blow your tree completely over!

Emotional costs and benefits can’t be shown with a decision tree the way numbers can. Whether you think an end branch outcome would be good or bad, you have to substantiate your belief with figures! However logical your decision might seem when you base it on the branch that promises to be the healthiest bet for your bottom line, any decision will only be as good as its implementation. Without emotional buy in from you and your team even the potentially brightest most profitable option is likely to wither and fail.

Used appropriately where finance is the deciding factor, brainstorming as many options as you can and estimating values with as much honesty as you can realistically muster, a decision tree can be very a very useful tool.

Beware that although they can appear to magically make the decision for you by showing up the best option, this can only be based on the information you’ve included. It helps to double check and question any biases and assumptions your tree may be rooted in!