Systemic Innovation: a discussion series

Systems failure and systems thinking

Professor John Seddon - 18.03.2013

Mid Staffordshire NHS Foundation Trust and successive tragic events in children’s services represent the tip of the iceberg of systems failure[1]. Recommendations from subsequent inquiries amount to no more than doing the wrong thing righter, for example improving inspection and accountability, rather than learning how to do the right thing.

While the Francis report on Mid Staffordshire acknowledges the adverse influence of corporate foci (targets and costs), it accepts these features as normal and necessary means of control, rather than seeing the consequences as signals that cause us to re-think our theories of control.

Horsemeat arrives in our processed food because the controls rely on form-filling. UK producers, far removed from the sources of production and focused on cost, will place the blame with suppliers rather than consider themselves to be responsible. In the same way health and social care workers feed the form-filling bureaucracy, which, in turn, becomes the focus for inspection, and, as a consequence, health and social care workers lose sight of their purpose, the de-facto purpose of the system having become avoidance of failing on the forms. It is the architect of the system who is responsible.

It is now well understood that people who need care or support are visited by a host of form-fillers from a variety of functionalised services who worry about meeting activity targets and protecting their budgets. Seeing what happens to people whose lives go off the rails, the best advice we can give them is, 'don't ask for help from the state' - for while the state intervenes in copious ways, it usually fails to help and can often make people worse off than before. The evidence from adult social care is that those in need have a long wait for support which, when it arrives, often fails to meet the need because the provision is governed through specifications and cost (price). Inevitably getting worse as they wait, people are then driven into care homes, the last place they want to be. Policy-making is now focused on shifting the costs to consumers, compounding the error.

The truth is counterintuitive: focusing on costs drives costs up. It doesn't take a rocket scientist to work out that we'd be better off if we could design a service that meets people's needs, quickly, effectively and once. If further evidence of the need for such a compellingly simple solution is required, consider that in health and social care 'failure demand' (demand caused by a failure to do something or do something right for the customer) [2] typically makes up as much as much as 80% of the total. When people re-present, they are treated as new episodes to be processed by the same inflexible industrialised service designs as before. While Jeremy Hunt believes a patient record could alleviate the lack of continuity, the greater priority is to design a system to provide continuity with care.

Pathfinders who have taken the latter route discover other counterintuitive truths about control in organisations. Crucial to successful service designs is thorough knowledge of demand - what people need, in their terms - and that becomes the crucible for everything that follows. They build a design that, placing individuals' wishes at the centre of service provision, is super-sensitive to people, their needs and their context. 

This sounds too good to be true: according to conventional management wisdom costs will soar as the normal means of control are relinquished. But studying the services as systems uncovers a different story: that it was the conventional controls that drove the system out of control. In the new design control is exercised by ensuring that the services exactly meet the demands placed on them. No longer monopolised by form-compliance, management's attention is instead focused on understanding demand and maintaining a body of expertise that will match that demand's variety; achievement of purpose being measured from the customers' point of view.

The extraordinary consequence of such action is that costs fall out dramatically, commonly 50% and as much as 80% in individual cases. This is a consequence of changing the system. The first step is to understand how the current system affects performance, which creates momentum for re-thinking our theories of management. The second step, designing services to absorb the variety of demand, then requires the emergence of different management practices underpinned by altered management philosophy. Systems failure needs to be understood with systems thinking, and systems thinking, if it is to have merit, must point to better ways to design and manage work.

From a system perspective these include rejecting targets in favour of measures related to purpose, cost-management in favour of value-management, standardisation in favour of designing to absorb variety, commissioning on unit price in favour of commissioning on the basis of cost, inspection in favour of prevention, and putting control where it belongs, in the work. Such challenges to convention need more than evidence (which is abundant); indeed, the way evidence fails to be the basis for reform is itself a system problem [3]. As those who have trodden the path would attest, the challenge requires being there - seeing is believing - something we cannot expect of the current policy-making machine. This machine, in turn, has to shift away from 'accountability' and instead place responsibility at the heart of our theory of control.

(The views in this blog are the author's own and not necessarily those of Nesta)

 

[1] Systems Thinking in the Public Sector: the failure of the reform regime and a manifesto for a better way, Seddon, J. Triarchy Press, 2008.

[2] Failure demand: demand caused by a failure to do something or do something right for the customer, see: Seddon, J. 2003 'Freedom from Command and Control', Vanguard Press.

[3] Seddon, J 2012 'The evidence is clear, but not at Whitehall', Public Servant, June: http://www.publicservice.co.uk/feature_story.asp?id=19921