Supplier performance measurement can be costly to implement if it is applied to all procurement unless there are sound IT systems which can help by capturing and analysing data originally input for other purposes.
All organisations need to be clear about why they want to measure supplier performance. Usually, it is to improve a supplier's performance (especially if the supplier has been performing badly), to ensure that timely action can be taken if the performance looks as if it will be inadequate (important if just in time is being used or on projects) and to provide a measure for helping to decide whether to continue to use the supplier or not (useful when compiling lists of potential tenderers).
Measurement should take account of the what is being purchased, the amount spent and the risk of failure of supply. A blanket approach is not recommended as it generates the same amount of information for non-critical supplies as for critical supplies. Even if this can be done cheaply, information for non-critical supplies can act as "noise" obscuring the information for important supplies.
Criticality can change. So it is important that any system is flexible so that it can be extended to suppliers as the need arises.