Thursday, May 20, 2010

The perils of QCBS

QCBS (Quality cum Cost Basis) is increasingly being employed for awarding various e-governance contracts (both Consulting and Software Implementation). It is a which tries to give weightage to both quality and cost. The weights are usually more for technical (or quality) and less for commercial (or cost) aspects of the bid (usually 60:40 or 70:30 or sometimes even 80:20).

The financial scores are usually calculated through a normalization process where the lowest bidder is given 100 and scores of all other bidders are normalized against this. For example – If A is the lowest bidder who has bid 20 crores for a project and B and C are the other 2 bidders with bids of 40 crores and 60 crores, normalized scores of A is taken as 100, B is taken as (100*20/40=50) and C is taken as (100*20/60=33.3).

Technical scores are calculated based on a variety of parameters such as experience in similar projects, quality of resources to be deployed (usually based on number of qualification, no of projects worked on, no. of years of experience etc.), write up or presentation on approach and methodology, average annual turnover etc with scoring being done for each factor separately and final technical score being a summation of these scores.

After calculating technical and financial scores, they are multiplied by respective weightages to find the total score and the vendor with the highest total score is supposed to be awarded the contract.

On the face of it, this system appears to be a very sound one and it gives higher weightage to the technical parameters which is as it should be. And it does work quite well as long as companies bidding for a project quote more or less in the same price range.

But if one of the bidders is a price player whose quotes are half or sometimes one third of the others (and this is not a hypothetical situation, I am not going to name the companies but I know quite a few cases where this has happened), then even if it is technically not very sound and manages to just scrape through in the technical cut off, it has a very high probability of winning the bid.

Lets take an example of 2 companies A and B which have bid at 20 and 40 crores respectively in a QCBS tender where technical to financial weightage is 70:30. A simple back of the hand calculation shows that if B has to score almost 22 marks (out of 100) more than A to make up for the deficit in financial scores. If it is a 60:40 instead of a 70:30, the margin goes up to 33. (See diagram below, unfortunately not very clearly visible, I wish blogger had a facility for attaching xls tables)The parameters of most technical evaluation are such that it is very difficult to score higher than anyone by such a huge margin.
















In the recent contract for UID software development, Mindtree, which won the contract, was the lowest (L1) bidder at Rs 19 crores. Accenture was L2 91 crores (see link). Just out of interest I had downloaded and read the RFP. The method of selection appears to be QCBS with 80% weightage to technical scores (quite high one would have to say) and 20% weightage to financial scores. Even with such a high weightage to technical scores, the large difference in quotes meant that Accenture had to score almost 20 marks higher than Mindtree (see calculation below) to get the project. That means that even if Mindtree had just met the Technical cut off of 70 (and I’m sure they must have done much better than that) Accenture would have to score more than 90 (out of 100) in the technicals to get the project. Obviously not that easy.

Moral of the story – if you know some of your competitors in the bid are price players who quote half or one third of what you do, rest assured that whatever magic you’ve planned for the technical, it won’t be enough to see you through.