In computer science and data management, 'null' is a special marker used to indicate the absence of a value or a non-existent reference. It is crucial for handling optional or missing data, but improper handling can lead to errors such as null pointer exceptions or data inaccuracies.