How to evaluate price in tenders is a topic that isn’t much discussed, but one that I have been personally interested in for many years. If we are bringing together price and non-price factors (quality, social issues, service, etc) in a tender evaluation, we have to convert prices (or whole life cost) into a “score.” And that is a complex subject, even more, complex when the prices submitted might span a number of different elements of the bid.
That seems to have been the problem behind the recent embarrassment for the UK’s Crown Commercial Service and their partners, UK Shared Business Services (which was developed out of the Research Council’s Shared Service Business). UK SBS now runs various tenders and frameworks on behalf of CCS, which is the central procurement unit now for all of the UK central government.
After legal challenges from unhappy suppliers, principally two of the leading firms in the construction consulting world, EC Harris and Turner and Townsend, the UK SBS framework for “Project Management and Full Design Team Services” has been scrapped. It would have covered project management, cost consultancy, architecture, and engineering, for use by all public sector bodies, but was already a year behind schedule.
Crown Commercial Services explain that they are “moving into direct procurement of services” as part of the reason behind the decision, but it seems clear that this was a very flawed process — with probably a list of suppliers and commercial arrangements that were sub-optimal. Piecing together the information that we have, it appears that the evaluation of the price in the tender was severely flawed. As Building magazine and website reported in September:
“T&T claimed rivals manipulated UK SBS’ scoring system – which was based solely on prices submitted for consultancy services in the South-east – by providing “very low prices for the South-east” and “very high upwards adjustments to those prices for regional work.” T&T said this was despite the fact regional work ought to be cheaper.”
So it looks like the scoring was based purely on prices for this one region, which allowed suppliers to “game” the system by putting in low prices where it mattered to the evaluation. That led to the situation where work would be cheaper in London than it would be in much poorer and lower-cost parts of the country – clearly nonsensical.
But this gaming is not uncommon in tenders. I saw another example recently where the bidders scored points both for the headline prices offered (day rates), and then separately for the volume discounts offered. But that meant you scored more points by offering a price of say €1000 a day with a potential 20% discount than if you offered €500 upfront but with a potential 10% discount!
The mechanics of scoring price was one of the big gaps we noticed when the UK government produced its Lean Procurement guide three years ago. There is much that is good in that approach, but it seemed to treat evaluation as a simple mechanical part of the process. It is not, as this case has proved. We would also observe that public procurement needs people with strong analytical skills, and a facility with numbers that would allow them to look at something like this and say “hang on, there is something wrong here”!
Going back to this UK failure, there is now the prospect of suppliers, whether successful on the original contract award or not, looking for damages to cover substantial bid costs. We suspect they won’t be successful in that, but this fiasco must call into question whether UK SBS is competent to deliver major contracting exercises for the government. On the evidence of this, they are not, and apparently, CCS will take the lead in the re-procurement of the framework.
And it is a warning for all contracting authorities – think very carefully about price evaluation and test out your mechanisms. One piece of advice to finish with. Find a good numbers person, and before you issue the tender, ask them to put themselves in the bidder’s shoes, and see if they can come up with a way of gaming your system!