While the merits of a lot of government policy can generate much debate, it is less controversial to suggest that the cost of public sector IT projects should be reduced as far as possible to get the best value for the tax payer. The Treasury estimates that IT costs the government around £16 billion per year. So what steps is the government taking, and how well are these being received? While there are a number of approaches that can be taken in the realm of business process, I’m going to talk about two ‘prongs’ of the government strategy that are based on the general approach to the technology itself and which are currently in the headlines.

(more…)

IPL welcomes the Institute for Government report “System Error: Fixing the Flaws in Government IT”. We believe that it will go some way to achieve its aims, however we have two principal concerns regarding its recommendations:

1. The report is overly focussed on ‘technology’ rather than ‘information’ – ie its recommendations are predominately related to technological issues. The technology is just the enabler; we need a far greater concentration on the information on which the technology operates. A high-level view of the UK’s information assets must be developed and a strategy formulated allowing better exploitation of information both within and across government departments. Unless and until there is a joined-up pan-government policy on the management of information, the country will not fully reap the benefits that IT can provide. Government CIOs need to remember that the ‘I’ in their title stands for ‘Information’ and not ‘IT’ or ‘Technology’.
2. As software developers, we welcome the report’s ‘agile’ recommendations but, as long-standing suppliers to a broad range of government departments and agencies, we know how difficult it will be for agile development to be exploited to full advantage in the public sector. Agile works well in the private sector where project sponsors are prepared to take ownership and  use their own agile-literate staff supplemented by external contractors, to deliver each iteration. In contrast, most public sector IT Departments rely heavily on IT Suppliers for software development, which will often involve customisation of software products owned by the supplier or third parties.

To use agile techniques effectively therefore, there needs to be a radical re-think in the way that the public sector manages and procures IT solutions.  This will require more client side IT-literacy and programme management capability. Use of agile will inevitably transfer substantial risk to the customer from the supplier demanding significant skill and understanding in order to manage those risks – and acceptance of those risks by the project sponsors.

Current procurement and contracting arrangements are not amenable to the application of agile techniques. The report proposes time & materials iterations, each with a value below the £105K OJEU limit. Realistically, this will not be practical for anything apart from a very small development (And it is our understanding that the breaking-up of large projects to get them under the OJEU limit is specifically outlawed under current legislation.)

Finally, an agile approach is not a panacea, suitable for all projects of all sizes. Indeed, there is a significant minority of projects for which an agile approach would be inappropriate.