Feature importane is an indicator of how much of an impact each feature(explanatory variable) had on the prediction result. Feature importance helps understand data sets and model structures easily, and helps users select features to develop more effective models.
[Click on the 'Feature importance' tab. Explanatory variables that had an impact on the prediction result are listed in order of their importance.]
In this case, duration, month, poutcome, contact had a significant impact on the prediction.