Systems Thinking in the Public Sector

Systems Thinking is one of those concepts that immediately made sense to me and helped me to understand what was wrong about the things that felt wrong about what was happening at work.

At the time I was on my way out of Lambeth Council in a cloud of frustration but took the opportunity to attend a conference on Systems Thinking run by Vanguard and hosted by their guru (if I can use that word positively) John Seddon.

A series of case studies helped to explain not on that another approach works but also why it works and why the government's top-down approach of target setting cannot work. This is possibly most obvious with school league tables which tell us nothing other than children are getting better at passing the tests.

Now Systems Thinking in the Public Sector is fully explained in a very readable yet thought-provoking book by John Seddon. It is so good I took it on holiday with me!

Why Performance Management does not work

I've been following the thread on Deming and Systems Thinking on the IDeA discussion board for over a year now and a recent post included the excellent THE GERM THEORY OF MANAGEMENT by Myron Tribus.

This is a very approachable description of what Systems Thinking is and why it is so hard for most managers to understand. I particularly liked this section as it reinforced my previous view that Performance Management is a destructive exercise.

"If you were the supervisor of these workers, what would you do? How would you go about improving things?

"I have presented this table to audiences across the USA, in Mexico, in Canada, in Australia, in the UK and I always get about the same reaction. People suggest a good talk with Eva. They propose putting Eva alongside Mary or asking Mary to help her. They propose to fire Eva. They propose to give Eva more training.

"One astute statistician at a meeting of the Royal Statistical Society in London even went so far as to observe that there was a 30 day periodicity in Eva's output and that might have something to do with things.

"After the audience suggests different cures based on the common wisdom, I explain to them that the numbers in the tables were actually generated by the random number generator in my computer. The flaws were generated and assigned to memory cells, to which I attached people's names. In other words, the faults were generated entirely by the system.

"In only two or three instances, out of thousands of people, did anyone suggest that perhaps the problem was in the system itself - that the system had been infected with the virus of variability and it was not the fault of the workers.

"In the last four years, only three people have suggested that we analyze the data in the table to see if we could compute whether Eva's results should be expected in the light of the variability exhibited by the system."

To try and simplify that as simply as possible, in order to understand the performance of a component (e.g. a person) in a system then you first need to understand the variation in the inputs.

A worst-case example is setting performance targets for dealing with that most variable of inputs, people. That is why it is impossible for Performance Management to work for jobs like teaching and social work.

A consultant's tools and techniques

A stray thought at work (these things happen!) has prompted this post on business modelling, knowledge management and systems thinking.

One of the side effects of working as a consultant for many years is that I have come across many modelling techniques and approaches to problem solving, and I think that one of my strengths is being able to pull the right tool or technique out of the hat when required. The problem is that much of the stuff that is in the hat is pants and should be left in there. Process Modelling is a case in point; I have yet to see a process model that was useful.

Over the years there have been a few things that I have come across that have immediately struck me as being "right" and in every case this opinion has been reinforced over the years. Three approaches that I have used again and again over the years are Data Modelling, Mass Customisation (Joseph Pine II's HBR paper), Knowledge Management and Systems Thinking.

The diagram here shows how I used the Mass Customisation to explain approaches to Knowledge Management at Lambeth Council. A key features of these different tools is that, just like in mathematics (my degree subject), it does not matter which of these approaches to a problem you use you will get the same solution. They provide a consistent description of the same domain.

Systems Thinking (a.k.a. Lean Service) is a more recent addition to my repertoire, boosted by a conference that I managed to wangle my way onto just before leaving Lambeth. The conference was given by Vanguard who are leaders in this area in the UK. Below is an extract from Vanguard's most recent newsletter which sums up why public service cannot get better under the current regime:

"Dear Mrs Kelly, I am moved to write to you today because I have just attended a presentation by [client name not reproduced here]. They have taken the systems approach to re-designing housing repairs and, like others, now achieve levels of performance beyond what might have been considered achievable if set as targets. In doing this work they have learned, as others have before them, that the statutory BVPIs are actually making performance worse. Attendees at the presentation - board members, senior managers and contractors - were concerned that, in the face of the evidence, your agents have failed to act. Like others in housing who have taken this route, these people now find themselves in the ridiculous position of having to measure and report things that they know are making their performance worse."

One of the nice things about learning about Systems Thinking was the realisation that this is the approach our school had been taking for years (e.g. designing for outputs and dealing with variations in inputs, i.e. children) which, for me, helped to prove the case for both Systems Thinking and our school.