Decision Trees

Decision Trees

1. Used when making decisions such as developing a new product or opening a new outlet. Lets them see what the cost will be, but the outcome is never certain.

   • Managers can work out a probability based on experience or past data
     • Expressed as decimal e.g. 0.6 for 60% probability
       • Total probability adds up to 1
       • The EV (expected value) of an outcome is the probability of it occurring, multiplied by the pay-off the business expects to get.

2. With Marketing: EV = (100k x 0.75) + (20k x 0.25) = £80k Net gain = £80k - 16K = £64K

3. Without Marketing: EV = (100k x 0.5) + (20k + 0.5) = £60K Net gain = £60k - £1k = £59K

4. Managers usually choose the course of action with the highest net gain.

5. The success and failure is added together using the probabilities because it shows the overall result of the decision.

6. Advantages

   • As it is numeric, it can be used as evidence to their decision instead of just saying 'this will increase sales'
   • Nice visual representation of the potential outcomes of a decision.
   • Allows managers to compare options quantitively and objectively, rather than going for the option they thought of first.
   • Useful in familiar situations where the business has enough experience to make accurate estimates of probabilities and benefits.

7. SQUARE REPRESENTS DECISION POINT

   • CIRCLE SHOWS ALTERNATIVE OUTCOMES

8. Disadvantages

   • Quantitative - based on numbers and ignore qualitative data (like employees' opinions). Businesses should take qualitative data into account before deciding.
   • Probabilities are hard to predict accurately. Estimated pay offs are assumed to be accurate. Dodgy info means decision is flawed too.
   • In reality there is a wider range of potential outcomes. e.g. decision tree only allows for success or failure instead of short term and long term success.

Images and files in the map