This is an example of a decision tree for knowledge discovery. Decision trees choose and display important factors influencing an outcome of interest. For demonstration purposes this decision tree predicts the likelihood that someone peformed automotive vehicle maintenance on a particular day as reported in the Bureau of Labor Statistics American Time Use Survey (unweighted sample). These results were generated using the data mining software Orange. Overall 2.6 percent of the respondents performed vehicle maintenance on a given day, but the percentage was 9.3 percent among males on weekends in non-metro areas. Both male and female respondents who work over a threshold of about 40 hours per week at their regular jobs are much less likely to perform vehicle maintenance as is evidenced by the splits on Work_Hrs_Week.
This is a demo using publicly available data. Applying a decision tree to your customer database can potentially yield insights into factors driving your business.