Public sector skunkworks

Last week I had a meeting with a council chief officer to discuss innovation within his authority. He characterised the council as having a 1980's culture, being very risk adverse, with a hierarchical structure and a command and control approach to management. He repeatedly used football analogies to make his point. In the innovation league, this council was mid-table.

How does an organisation such as this set about becoming more innovative?

Another chief officer in the same council suggested an approach that she has now twice tried in two different councils. This is to set up an innovation incubator, a sort of research and development unit for public sector innovation. This incubator would operate in a different way to the rest of the council. It would be more entreprenural. It would have its own culture, distinct procedures and be empowered to experiment with new technologies to the council.

This is the skunkworks approach. The term was coined at Lockhead, the US aircraft manufacturer. In the 1940s, the main factory at Lockhead was assembling bombers for the British forces. They were approached by the US military to develop and build a jet fighter capable of combatting the perceived growing threat from Germany. A small team was put together, working in secret, separate from the main organisation, with the authority to take risks and make some decisions without referring to the line of command.

According to Wikipedia, skunkworks is "widely used in business, engineering, and technical fields to describe a group within an organization given a high degree of autonomy and unhampered by bureaucracy, tasked with working on advanced or secret projects".

The public sector skunkworks was an interesting organisational innovation to observe. In its three-year life it gained a number of national accolades for its technological innovations. But its distinct culture and private-sector approach caused many tensions within the council which eventually led to its demise. The Lockhead skunkworks has survived for many decades. Why did this public sector version fail so quickly?

I'm on the pavement thinking about the government

Once again today there is another story about the failure of public sector innovation.  This time it is the national offender management information system which was set up to provide 'end-to-end offender management'.  This aimed to share information between prisons and the probation service, replacing separate stand-alone systems.  At its inception in 2004, the estimated lifetime cost of the system was estimated to be £234M.  By 2008 the expected total cost had more than doubled.

The National Audit Office review, published today, highlights the following failures in this project

1.  Senior service management did not monitor the project sufficiently.

2.  There was no clear governance structure in place.  Roles and responsibilities were blurred.

3.  Poor programme management - the lack of an overall project plan;  poor change control and inadequate financial control

4. The need to perceive major projects as IT-enabled business change rather than technical IT projects.  The 42 probation services all had individual ways of working.  Trying to accommodate these led to requirements drift.  Processes should have been standardised and simplified.

5.  Poor supplier management.  This is an outsourced development, but it was not tendered for.  The development progressed on a time and materials basis, racking up the costs as requirements grew.