Who goes on holiday to Sweden?
This worked example builds a Decision Tree to identify the characteristics of customers who have been on holiday to Sweden.
The Decision Tree will enable us to identify the characteristics which distinguish Swedish holiday-makers, and will turn them in to a selection rule which you could then apply to a prospect database. For example, people who go on holiday to Sweden might turn out to be, either:
- Young males of any income
- Older females who are affluent
- Middle aged people of either gender
Setting up the Decision Tree¶
The main stages involved in setting up a Decision Tree are:
- Drag the Decision Tree icon from the Selection Tools to the Workspace
- Set the Analysis Selection
- Set the Base Selection
- Set the Dimensions
Having built the tree you can then examine the output at a high level, or go in to detail to understand the numbers involved and how each split has been made. Graphical views make the process of understanding the tree easier.
The Decision Tree will enable you to identify your best customers. If need be you can refine the Decision Tree which has been built
The Decision Tree will enable you to select your best customers, using the selection rules to plan a Marketing Campaign or to just gain a better understanding of your customers.
Once you are familiar with the basic principles of the Decision Tree, you can experiment with different build options to tailor the Decision Tree to your data.