228 – Frog or bike. Does it matter?

1720 words (6 minutes reading time)                                                                Lancing Farrell

thinking frog

Colin Weatherby and Tim Whistler have made some interesting points in their contributions to the discussion about the Vanguard Method. Clearly, Colin’s colleague has had some success in using it and has been able to identify unique features of the Vanguard Method. In contrast, Tim has raised some legitimate concerns, especially from a local government perspective. I have spent some time thinking about both points of view and reading some of the material mentioned. Here are my thoughts for what they are worth.

Services are complex and the interactions between the parts, especially when people are involved, is important. Experience says that changing one part of a system does often have unintended consequences elsewhere. Even using the bicycle system as an example, changing one worn-out tyre isn’t going to improve the whole bike if the other tyre is also worn out. A blow out is still possible. Upgrading the group set to a 500g lighter (and much more expensive) set isn’t going to make much difference if the rider is 10kg over weight. Successful cyclists understand what they are doing as a ‘system’ and coordinate their effort across all aspects of their cycling.

‘Frog’ systems add another degree of difficulty. Sometimes, that difficulty can’t be seen or understood until you start to change the system. This is why ecologists intervening in natural systems often use ‘adaptive management’ approaches – e.g. trial and error – to make small changes and monitor the effect before committing to do more. They have adopted a structured and iterative process in the face of uncertainty. I know organisations are not ‘natural’ systems but human nature does apply and people don’t always behave predictably.

On the basis of this, I would support systems thinking as a way to understand an organisation and implement change. It places the emphasis on the ‘whole’ rather than the ‘parts’. Despite Tim Whistler’s concerns, risks must be taken to change a system. The trick is to manage those risks. Even when changing a bicycle system success is not guaranteed. There are many examples of organisations that have set out to implement structured, planned and documented changes that have not achieved the intended outcomes. Even having multiple change processes and coordinating them is not straightforward. It might be what we are used to doing, but that doesn’t mean we should continue if there is a better alternative.

At this point, the differences of the Vanguard Method become relevant.

Anything you do has to start somewhere. In a service, it makes sense for it to start with a person receiving the service and their purpose in using it. There doesn’t seem to be any point in giving someone a service that isn’t what they need. I am not saying they should get what they say they want, but unless they get what they need to solve their problem, they will not be satisfied. Then they must either put up with what they get or they will seek to solve their problem by re-presenting or going elsewhere. This is the source of the failure demand identified by John Seddon. It is more likely to occur if someone other than the customer has determined the purpose of the system.

This is an interesting point and raises the question of why the work is designed the way it is. If it is not to fulfil a customer purpose, whose purpose is it intended to fulfil? You might argue that it is the role of leaders to determine the purpose, but does that mean it should serve their purpose? It is a common criticism of public sector organisations that work is designed to suit the workers. Even designing the system to suit external regulators or to satisfy regulatory obligations or reduce risk is not going to work unless it also fulfils customer purpose. It is customers who create the value demands in the system.

If you study enough customers using a service, there will be common and predictable problems to be solved. This is the ‘requisite variety’ and the value demands that should form the basis for designing the system. It is not ‘standardisation’ because the system will be designed to respond effectively to what matter for customers in solving their problems. And it doesn’t mean that it won’t change over time. Once you have worked out how to deliver a service to meet value demands, you can continuously change it in response to changing demands. If that happens, the periodic service transformations we use to re-set services when they get out of step with customers won’t be required.

The worst thing that can happen is that the service design responds to unpredictable problems or exceptions. This is how additional steps get added into work flows and much of the avoidable waste (i.e. work done that doesn’t add any value for the customer) finds its way into systems. This leads to Tim’s concern that the Vanguard Method won’t deliver public value.

Mark Moore defines public value as the “collective view of the public or community about what they regard as valuable with regard to the use of public money and authority”.

Moore has placed public value on a continuum from private to public value. This is discussed in more detail in an earlier post and the fundamental concept is illustrated in this diagram.

Moore degrees of publicness

Allowing people to define the system from the viewpoint of their problem to be solved is clearly a private value situation (i.e. an individual seeking material well being). This is an unavoidable situation because people don’t like standardised services. We know that. They want choice to ensure their needs are met. My favourite example is to imagine how you would feel if doctors treated each patient according to the illnesses prevalent in their post code, rather than examining them for their specific illness. I can’t imagine anyone thinking that would be helpful or acceptable to the patient.

 Staff have regularly reported that what starts out as a typical customer transaction (i.e. I want something and I am paying you to give it to me) often ends up with the customer taking a more ‘citizen-like’ view of the situation and the council’s responsibilities, which results in the solution to their problem becoming one that is better for them and their community.

However, if, as evidenced above in the advice given by Colin Weatherby’s colleague, a discussion that starts with private value can move along the Moore’s continuum to end up with a public value outcome (i.e. both individual and collective valued states are achieved), surely it is public value. In fact, it is likely that all demands have elements of both private and public value. An example used in a previous post to illustrate the transition of value from private to public has been the planting of a street tree outside a house in a street. The resident living in the house gets value in beautifying their street, as does the broader community in reducing the heat island effect.

The issue of performance measurement is always contentious. If measures are not used to understand whether services are effective as they are delivered, how can they be improved? Relying on surveys of customer satisfaction (which seems commonplace in councils) or waiting until external comparison reporting shows you have dropped down the ‘league table’ of councils, is inadequate. I can’t imagine successful private sector service organisations waiting for that sort of feedback. In reality, there probably needs to be both sorts of measurement and reporting – leading measures that help improve systems and services, and lagging measures that reassure decision makers and the community that needs are being met and problems solved.

The link made by Tim between todays services and the services needed tomorrow is an important one. There is no point in putting improvements in place today that solve a problem and then just waiting until a new problem emerges before doing further analysis and change. We know services are dynamic and that customers and the variety they bring to the system will change over time. Why not set up a system capable of changing continuously in response to demands?

I think the main reason this doesn’t happen often is that management hasn’t worked out how to do it effectively. It also challenges the traditional ‘command-and-control’ mindset of leaders. The idea of a service catalogue is one that I have long supported (I have posted on it before). My rationale has been that no one knows with certainty what an organisation does, or does not do, unless you tell them. Write it down. Having thought more about the Vanguard Method, I can see that a rigid catalogue of services and service levels is not the solution, particularly if it is developed ‘inside-out’ by people within the organisation. The services offered need to be those required by customers or they are irrelevant.

This leads to Tim’s final point, and one that is addressed partially in Colin’s post – what happens to leaders? Leaders must change if the Vanguard Method is to be successful. It is probably the greatest threat to its success. They must change their thinking first because that will lead to a change in their behaviour. The power of what leaders do is well documented. The blog Thinkpurpose.com has a popular post about systems thinking called the ‘7 reasons you shouldn’t touch systems thinking’. The seven reasons are the frustrations that occur when workers are exposed to systems thinking but management is not committed. This seems to be a real risk.

It is a risk for leaders to adopt systems thinking and introduce a method like the Vanguard Method to change their organisation. Leaders have accountabilities that are designed around ‘command-and-control’ thinking. Many leaders have been successful in a ‘command-and-control’ environment – they are accustomed to deciding what will happen next, creating priorities and allocating resources. People rely on them to do it. However, leaders that continue with reductionist management thinking and practice will defeat the Vanguard Method or any version of systems thinking.

They have to change first.



191 – Essay No. 3 – Local government and systems thinking.

Posted by Lancing Farrell                                                                              1250 words


systems thinking checkland

Systems Model, Peter Checkland, 1981.

Systems thinking has featured in a number of posts (see  Some types of thinking observed in local government,  Classic paper: ‘Forget your people – real leaders act on the system’. John SeddonApplying the public value concept using systems thinking in local government). As someone with an interest in systems thinking I felt it deserved some discussion in the context of local government using ideas gathered from Peter Checkland, Alistair Mant and John Seddon.

The cover of Peter Checkland’s book ‘Systems Thinking, Systems Practice’ says that it is about the ‘interaction between theory and practice of problem solving methodology’, as derived from a decade of action research. It is a seminal text on the ‘meta-discipline’ of systems thinking.

Checkland has set out to ‘develop an explicit account of the systems outlook’ and, based on that view, to ‘develop ways of using systems ideas in practical problem situations’. The book is about the ‘use of a particular set of ideas, systems ideas, in trying to understand the world’s complexity’.

“The central concept ‘system’ embodies the idea of a set of elements connected together which form a whole, this showing properties which are properties of the whole, rather than properties of its component parts.”

Systems thinking is centred on the concept of ‘wholeness’. In some ways it is a reaction to the classical scientific method, which emphasises ‘reducing the situation observed in order to increase the chance that experimentally reproducible observations will be obtained’.   Eliminating variables in order to study something to determine cause and effect is useful but ultimately limiting in complex systems where the interactions of all variables matters.

Systems thinkers have called it ‘organised complexity’ to describe the space between ‘organised simplicity’ and ‘chaotic complexity’. In some ways it is seeking to understand the simplicity that exists on the far side of complexity. In a nutshell, systems thinking is concerned with organisation and the principles underlying the existence of any whole entity.

Australian management author, Alistair Mant, describes two types of systems in his book ‘Intelligent Leadership’. He calls them the ‘frog’ and the ‘bicycle’ systems and he believes that leaders need to be able to distinguish between the two systems when applying systems thinking and directing a change and shift in systems. He sees ‘pointing systems in intelligent directions’ as one of the critical leadership responsibilities.

Frogs and bicycles are metaphors for different kinds of systems. The essential difference lies in the relationship of the parts to the whole. A bicycle can be completely disassembled and then reassembled with confidence that it will work as well as before.  This is not possible with a frog. Once you remove a single part the whole system is affected instantaneously and unpredictably. Furthermore, as you continue to remove parts the frog will make a ‘series of subtle, but still unpredictable, adjustments in order to survive’.

“This sort of system, at a level beneath consciousness, wants to survive and will continue for an astonishing length of time to achieve a rough equilibrium as bits are excised – until it can do so no longer. At that point, again quite unpredictably, the whole system will tip over into collapse. The frog is dead and it won’t help to sew the parts back on.”

This is a salient warning for those planning to intervene in a system without due care. Mant believes that most big organisational systems contain bits of frog and bicycle systems.   He says that the bicycle parts can be hived off and reattached in a new way without harming the overall system, but that the frog parts are really the core process.   In a way, he is describing systems at two different levels – the component-level (bicycle) and the system-level (frog).

Checkland says that ‘any system which serves another cannot be modelled until a definition and model of the system served is available’. This approach should prevent action on component ‘bicycle’ systems occurring before the potential implications for the whole ‘frog’ system is understood.

In common with Checkland, Mant holds the view that systems are complex and need to be considered as a whole. The systems model developed by Checkland (shown above) illustrates the process of systems thinking. He says that it starts with a ‘focus of interest’ or set of concerns that exist in the real world. This could be a problem or something about which we have aspirations. This leads to an idea. From that idea two kinds of theory can be formulated:

  1. Substantive – theories about the subject matter.
  2. Methodological – theories about how to go about investigating the subject matter.

Once theories exist, it is possible to state problems not only as they exist in the real world, but also as ‘problems within a discipline’. For example, engineering, chemistry or town planning. All of the resources of the discipline (i.e. previous problems, its paradigms, models and techniques) can then be used in an appropriate methodology to test the theory.

The results of this test, which involves action in the real world (i.e. interventions, influencing and observation), then provides case records of ‘happenings under certain conditions’.   These are the crucial source of criticisms that enable better theories, models, techniques and methodologies to be developed. It is the improvement loop.

For those proponents of the Vanguard Method this might be starting to sound familiar. John Seddon has developed an application of systems thinking in organisations that has demonstrated its value in improving organisational systems. The Vanguard Method has the following steps:

  1. Check
  • What is the purpose of this system?
  • What are the types and frequencies of customer demand?
  • How well does the system respond to demand?
  • What is the ‘flow’ of work?
  • What are the conditions that make the system behave this way?
  1. Plan
  • What needs changing to improve performance?
  • What action could be taken and what would we predict the consequences to be?
  • Against what measures should action be taken?
  1. Do
  • Take the planned action and monitor the consequence in relation to the purpose.

It features systems thinking in the need to understand a concern in the real world (what are the demands?), develop a theory or approach to improvement (what needs to change?), and then seeks to implement that action and monitor consequences (how can the approach be improved?).

“The outcome of studying the work in this way is a system picture that puts together everything that has been learned and which illustrates the dynamics of the particular service.” John Seddon.

Mant says that most complex systems containing and serving people have ‘natural properties’. Effective management aligns itself with the natural flows and processes to help them along – like a leaf floating on water running in a stream that naturally takes it to its destination. Bad or dogmatic management fails to recognise these natural properties and attempts to ‘shoehorn the system into shape’ to meet externally determined priorities. This has been identified as a problem for public sector management in a previous long read post.

Local government has spent too long looking out of the window hoping to see a ’business-like’ way of managing that will solve all of its problems, rather than having the confidence to work out what is needed from first principles. It is vulnerable to the next externally imposed management fad.

Seddon is particularly harsh in his judgement of public service reform for the past 35 years in Britain. He describes the contribution of each Prime Minister in some detail. Overall, he paints a picture of political interference and the projection of a narrative that has primarily focussed on reducing costs, yet costs have increased. This is principally because they have failed to understand the system. The complexity of systems containing and serving people has been overlooked.

Checkland, Peter, 1981. ‘Systems Thinking, Systems Practice’.

Mant, Alistair, 1997. ‘Intelligent Leadership’.

Seddon, John, 2014. ‘The Whitehall Effect’.