As explained in the section 2.2.1., the K-Means algorithm is an algorithm with a strong random component: each time you run the K-Means algorithm, you find a slightly different segmentation. You should thus run the K-Means algorithm several times and “have a look” at several different segmentation. Thereafter, you can select the segmentation that has the best interpretation from a business-point-of-view. You can easily interpret you segments using the charts and tables accessible with the , and buttons.
An easy way to create many different segmentations is to click repetitively on the button. Each time you click on the button, a completely new segmentation is computed and shown.