Provided is a failure predictive system, including a collection unit that collects information with respect to a job implemented in a monitored apparatus and information with respect to a job to be implemented, a calculating unit that calculates a probability of a failure occurrence in the monitored apparatus based on information with respect to the implemented job and information with respect to the job to be implemented which are collected by the collection unit, and an output unit that outputs information based on the probability when the probability calculated by the calculating unit is over a threshold value.