performance target, friend or foe?

I subscribe to Stacy Barr’s blog on performance measures and the other day she posted an entry on targets. Since I am working towards removing targets all together I posted a reply.  

Stacey’s post is available on

My reply is published below and I’ll be interested in Stacey’s reply.

Hi Stacey, I’ve commented on this topic before and in my view targets are a complete waste of time – we should not become friends with them we should abolish them. The negative consequences of targets and a target setting culture are quite astonishing.

What is wrong with just measuring performance with the intent of learning and improving? What value does the numerical target add to this, especially since it will be an arbitrary number anyway? Any number you put up could be questioned, we need to improve by 10%, why not 11% or 9%? – the real question is by what method will we improve our performance? 

Even in safety with Zero harm targets which on the surface sounds admirable, the danger is that people do not report things as it will ruin the reporting.

Targets motivate people – if so where is the evidence for this claim? I’ve seen much evidence to suggest the contrary, targets demotivating staff and creating a de facto purpose to reach the target regardless of broader impact, all sorts of creativity comes into play to manipulate data to reach the target. A recent report from House of Commons Public Administration Select Committee (PASC) into allegations of police mis-recording of crime statistics highlight this.

Simon Guilfoyle’s blog and link to the report is available on

I’ve borrowed this marvellous image from Simon’s twitter feed.


Targets sets direction – surely a well defined sense of purpose already does this. If people do not know why they are doing something that is a critical failing of leadership.

Stacey, you know your statistics so when I say that there are only three places a target can reside on a control chart you know what I talk about (assuming a stable process). It will either be below the LCL (in which case you will always meet the target), it will be in between the UCL and LCL (in which case it will be hit and miss due to common cause), or it will be above the UCL (in which case you will never reach it with the current method).



Your example of reducing the time from 24 hours to 12 hours is an interesting one, why is the average time important? A couple of complex cases and your average can blow out. You have spoken before about the need to shift the process capability and that’s a way to understand if any real improvement has been achieved. The value is as you know not in the average as such.

Targets and target setting is the thinking of the command and control manager in a organisation where decision-making is separated from work. As the complexity of organisations increase and more and more work is knowledge work, managers have had to abandon the attempt to control the individual activities of their staff and stop using target-setting to try to control the organisation.