It is common practice that in teams working in the field of natural sciences all group members manage their primary data in highly individual systems. The consequence of this is that the data are usually lost shortly after the scientific results have been published or that they lose part of their value, as significant background information can no longer be found. To solve this problem in a simple way, we present a basic procedure that allows us to uniquely identify scientific data and their history at any time. We describe which requirements such a procedure has to meet (proper documentation, unique identification, and easy backtracking of the individual operations) and discuss on the basis of a timestamp approach how such a procedure can be integrated smoothly into the traditional scientific work process and the existing IT infrastructure of a team. It does this by using established processes and developing them into a systematic information management covering both electronic and analogue media.