The defining feature of these models is that the dependent variable is the time until an event occurs. They were originally developed for mortality events–understanding the relationship between predictors and how long after a treatment patients survive.
But it turns out you can adapt them to many other time-to-event outcomes, such as customer churn, restaurant closings, successful completion of training for guide dogs, length of unemployment, defoliation of plants, and metal fatigue. The outcomes do not necessarily have to be “bad” events or anything to do with surviving.
Join Steve Simon as he introduces you to some fundamental tools and concepts within survival analysis.