Evaluating Tenders – Scoring and Manipulation

Tuesday, July 12th, 2016

EVALUATING TENDERS – SCORING AND MANIPULATION

The evaluation of tenders is a highly professional process, aimed at ensuring there is a credible approach to appointing the right supplier. The process should be visible, known to the bidders and beyond reproach.

Our research has led us to another questionable approach. It relates to the provision, installation and maintenance of telephone equipment to enable prison inmates to make calls. The requirement was tendered and the award criteria was first of all that the winning offer would be ‘that which offers the state the most money.’ This was changed to ‘whose offer has the highest point total.’ A pricing table was included in the invitation to tender. There were 1000 points available in total. Of these 550 points were for the ‘Commission Rate’ which was the amount of income the provider would pay the state. There were then 450 points for six categories (shown in the table below).

Maximum Points Category
200 Rate in $ per minute – domestic
  50 Rate in $ per minute – international
  70 Rate in $ per pre-paid account set up fee
  80 Rate in $ per fund transfer into pre-paid account
  30 Rate in $ set up charges per call – domestic
  20 Rate in $ set up charges per call – international

 

The audit report includes the observation that tailoring the evaluation in this way meant that the winning vendor who bid $0 for 4 of the six categories would receive maximum points in those four categories thus assuring at least 400 points of the 450 available. The way pricing was structured and since there were no caps on rates specified in the invitation to bids, a vendor could have bid any amount ($20 per call, $50 per call, etc) under the set up charges and still had the highest point total to be awarded the contract.

The skill that was required in this situation (from a buying point of view) was to determine the cost of a pre-determined length (for example, a 15 minute call) by combining set up fees and per minute charges and then assign points proportionately. This would prevent a vendor from placing all of its costs into a single category solely to maximise points.

This should make all buyers think. Designing a tender evaluation model requires the different scenarios to be forecast and theoretical evaluations undertaken to spot potential flaws in the evaluation model.