Whether in the defence or other domain, computer-based tools are used as aids to the systematic evaluation, documentation and comparison of competing tenders.
These tools often employ quantitative techniques to determine ‘value’ indexes as quantified measures of the worth to the buyer of an offer and to use the indexes for the direct, ratio comparison of offers. The process is sometimes taken further to calculate ‘value for money’ indexes, by the division of the value indexes by respective tendered prices, which may be used as an additional basis of comparison.
This paper contends that value indexes and particularly value for money indexes are more limited in their power than given credit for by proponents. Specifically, this paper contends that:
- value indexes are capable only of ranking competing offers and do not permit the valid comparison of offers as ratios, ie value indexes can say that one offer is better than another, but cannot say how much better;
- a value for money index is an undefined quantity, devoid of any real meaning and invalid for use in comparison of offers;
- unless special precautions are taken, some evaluation models may not produce reliable results, in that rankings of offers may change depending on data values applied to model parameters; and
- that value indexes and tendered prices should only be used as separate data inputs to qualitative (narrative) analysis and comparison of offers.
Thus, this paper explores the validity of quantitative indexes and their limitations in tender evaluation.
It does not challenge the benefits of these models in the systematic evaluation and documentation of offers, which can be a daunting task in its own right, nor discuss other aspects necessary to complete and useful evaluation models.