It's performance management Jim, but not as we know it...

I have read two rather excellent pieces this morning about managing and improving performance in the public services.

First I came across this piece by Mike Ledwidge entitled "Why has it all gone wrong within our public services". It is a bit of a polemical rant containing some ideas that I would not support* but its main thrust that the public services have been failed dismally by politicians and managers who think performance can be managed by measuring outputs is well made. To cite one paragraph (but do read the whole article):

You CANNOT performance measure a ‘complex system’ by outputs. Now if you do not understand EXACTLY what that sentence means let us hope you are not involved in anything to do with the management of our public services. Sadly we now have thousands of senior public servants who think they do know what they are doing with targets and measurement, and clearly they don’t. Complex systems have more than one purpose. If you measure the police on arrests and detections, any prevention they do will muck that up. If you ‘performance measure’ on crime reduction, officers will find ways to not record crimes. The awful tale of the rape unit in Southwark trying to improve their stats is an example of the result of government pressure and targets. 

And then later, @TheCustodySgt pointed me towards an excellent piece by @SimonJGuilfoyle entitled "Panic!"  In his article, Simon uses his long experience of such matters to highlight how managers often lurch into action based on an erroneous understanding of performance variation. Again, please read the piece as it contains a delightful cartoon which makes the point very clearly:

The post is about the unintended consequences that can occur when managers draw erroneous conclusions about data

As regular readers know, this is a subject I have mentioned before in several places. My last post on here has many links.

So for the uninitiated politician and manager charged with the responsibility of improving public service performance and getting quarts out of pint pots... here are some pointers:

• Achieving social outcomes (the improvements in society that we pay the public services to produce) is mighty complex: don't even think you can boil things to simple linear or transactional 'customer' relationships!
• Everything varies: the weather, leaves in the forest, need for social care and disturbances on drunk Saturday nights...
• But, there are patterns in these variations which need to and can be understood (well mostly): public services need to be resourced and organised around these variations
• Measurements & targets change that which they are measuring and targeting (and not just in the way that a watched pot never boils!)
• As Deming famously said "drive out fear": if your system of performance management contains even a wisp of fear, people will do weird and unexpected things that are not what you intended

I could go on...

But please, just read some work by Deming, Checkland, Ohno or Seddon. And please (please!) stop wasting precious public resources on fluffy, vanity systems of performance management that are mostly "sound and fury signifying nothing!"


* e.g. the comment "At one stage we were 20,000 teachers short, and some have been replaced by people who, like some doctors, are not easy to understand" which is an unnecessary xenophobic swipe, it seems to me