I see that Dilbert has picked up on Six Sigma. Or, to use my deliberate slip of the toungue (to which I confessed in a response to a recent comment): Six Stigma.

Now from my perspective Six Stigma is very different from BPR other than that its execution seems more determined. As a result of this it may be more difficult for the informal networks in the organisation to mitigate its excesses. That itself is an interesting aspect of organisational life. A lot of stupidly survives simply because good people use their their networks to prevent very bad things happen. As a result of which the negative feedback to the system designers is suppressed. You could argue that a good old fashioned work to rule would demolish many a fancy management initiative. As it is, especially in the public service ,people find ways to make the system work despite itself.

Now the more rigid process based techniques are of course useful; where where you have a degree of stability and where the measurement systems reflect actual performance and are non-gamable. This follows the principle of bounded applicability that is at the heart of Cognitive Edge methods: most methods work in context, but few if any are universal in their application.

It is noticeable that most of the more explicit methods around all originated in manufacturing. Such systems are closed with repeatable outcomes so its not surprising that explicit methods can be developed and propagated with success. The problem happens when you shift the context, and apply methods defined for closed systems to open ones. BPR started to go wrong when it was applied to services, a contextual shift not supported by the original data. However techniques such as BPR and Six Stigma are very attractive to senior management as they appear to allow managers to control things by measuring outcomes. It is also interesting to observe that over measured systems lead to poor management. The measurement system can be used as an excuse to abrogate responsibility for exercising human judgement and taking responsibility for the consequences of those judgments. It is a lot easier to hide behind a set of figures.

Of course (and this is the title of this post), if people want to live in a neat, tidy and ordered universe then they will find it difficult to accept that such measurements can often produce the opposite of what they were looking for. I have previously commented on the stupidity of outcome based measures . I may have related my own experience of trying to secure an appointment with a doctor more than 48 hours in advance of the said appointment. The British goverment had created a target for all patients to be seen within 48 hours and most doctors have achieved the target by a simple process change. It is a great example of the gaming behaviour which is endemic in over measured systems. I now learn that instead of realising this absurdity, the Government intent to compound their error with an ill thought out and suggestive satisfaction survey which will set more explicit targets, with more money linked to those targets and which will be gamed in turn. Its an illustration of the principle

When will this absurdity stop? How much money is being spent on measuring the un-measurable and loading target on each target as each new attempt fails or is gamed to produce the desired outcome as the expense of common sense behaviours?

< Prev

Serendipity strikes again

Thanks to Tom Quiggin for this example of innovation. It matches the history of science ...


Further Posts

Next >

Torture & Anthropology

There has been an interesting debate in the Savage Minds blog on a recent American ...


Further Posts