back to main menu
all posts

At the end of last year, the UK’s National Audit Office issued a very useful document titled “Commercial and contract management: insights and emerging best practice”. We did provide an initial overview of it here, and now we are into a more detailed review of its content and findings. For each area the NAO has covered, we will look at their content and then give any additional analysis or thoughts we would add to the mix. Today, we will take a look at number 12 of the NAO insights (as they call them) –  “Design performance measures that work”.

Without relevant and workable measures, NAO says, “business outcomes may fail or perverse incentives may be created”.  And in their experience, they “rarely” see measures working as intended, which is a real concern of course. NAO also sees clients who are happy when measures are not being met (which suggests the measures weren’t appropriate) and other clients unhappy when they are met (the same problem!).

So, “we need to question not only whether the performance regime is right (does it incentivise the right behaviours, penalise poor performance and ensure that government gets what it wants?), but whether government knows what performance is and can manage contractors to get what it wants”.

That is a very fundamental point – if you don’t know what you really want from a contract and supplier, how can you define the right mechanisms? Measures should be aligned with the desired overall business outcomes, reflect an appropriate allocation of risk, not be too complicated, and financial incentives must be big enough to have an impact. Measures can become out of date during the contract, and should be amended where that is required; but without altering the balance between risk and reward if possible. 

The case studies here show both good and less good practice, or in one case both. The NAO found that the Department of Work and Pensions’ Work Programme initially used flawed measures that meant even the worst performing suppliers received incentive payments. But DWP “changed the way that it approached performance measurement two years into the scheme in response to the difficulty of monitoring and enforcing minimum service standards” and that improved matters. 



Public Spend Forum Comments

This is another heading that tends to make readers think “well yes, that’s fairly obvious” and yet there is plenty of evidence that appropriate performance mechanisms, incentives and penalties, have not always been in place even for the biggest and most important public contracts.

But we have sympathy for buyers here, as this is a very interesting but also very challenging topic. It has seemed to us for some time that “incentivisation” as a topic deserves to have its own MBA, as it is so complex and intellectually challenging. But incentives drive behaviours, in individuals and organisations, so getting the right contract measures is vital. When we look at some of the most complex service contracts that suppliers re now delivering, such as offender rehabilitation, getting people into employment, or complex health and social care issues (dementia care or drug treatment, for example), it is easy to see how critical and yet how difficult it is to design the measures that really focus suppliers on the desired outcomes.

This area therefore needs a real effort in the public sector in terms of education, communication and even basic research, we would suggest. There is good practice around, but it is not always spread as widely as it should be.

Finally, we have to mention politics again. As with other factors we have discussed already for the NAO report, choosing measures can become a politicised process. Politicians want to show their motives are working, hence measures can sometimes be skewed in that direction. Do you measure offender rehabilitation in terms of no re-offending for 6 months after release from prison or 2 years? If the Minister wants to claim credit, then it is obvious which will give the “best” apparent result! 


Market and Supplier Intelligence
Powered by AI-MITM
Explore More

No Comments

You must be logged in to post a comment.