Glued-shut IT wallets hindered UK govt's programmes – study
Viewing big initiatives as tech projects not helpful, say Brunel researchers
Efforts by the previous UK government to rein in lavish Whitehall technology spending caused more harm than good in some instances.
That's according to a new academic paper, titled Identifying the critical success factors for major government projects that incorporate IT or “digital” developments. It builds upon two independent probes into the disastrous Rural Payments Agency's Common Agricultural Policy (CAP-D) system and the Universal Credit fiasco.
This latest study's author, Professor Vishanth Weerakkody, said those two projects illustrate how changes to the governance and management introduced since 2010 "have significantly increased the risk of failure, rather than reduce it as intended."
In 2011, former Cabinet Office minister Maude imposed government-wide spend controls on recruitment, consultancy, property, advertising and marketing, and information and communications technology (ICT). All new ICT spending above £5m was subject to Cabinet Office approval, and anything over £1m for software systems.
Those reforms were widely viewed as having saved the government significant amounts of cash by stopping costly projects. However, the paper noted that in relation to some projects, the spend controls hindered rather than helped.
"Many of the problematic issues relate to interventions introduced by the Cabinet Office after 2010 that were intended to improve the execution of such projects, but in fact made matters worse," Weerakkody noted.
The Brunel University professor of digital governance added that while spending controls on IT have been perceived to be useful as a means of scrutiny and challenging IT spend is important, government programmes ought to be viewed as means of implementing policy rather than as purely IT projects.
"There were a lot of very sound and usable interconnected processes that were developed over 10 to 20 years. Once you rip out or add in components without understanding the whole system, you are going to break the system."
He said: "There needs to be a real fundamental rethink about how all these processes fit together, and repair the corporate loss of memory."
The Treasury versus the Cabinet Office
Weerakkody argued that the Treasury, not the Cabinet Office, should have final say as to whether a project goes through, based on scrutiny of a proper business case, including the IT element, for the policy implementation as a whole.
He added that the recent Public Accounts Committee report on the Cabinet Office's shared services centres project – which has failed to deliver value to the tax-payer – had similar issues in terms of proper oversight.
"In that case, there was also a shambolic lack of governance and no business case." Weerakkody said that while Universal Credit was a mess in multiple respects, the apparent misapplication of agile to the project created further complications.
The report noted that although the Treasury process for approving business cases is well-established, in 2011 a separate IT spend control process run within the Cabinet Office became mandatory for projects over a certain financial threshold.
It said: "This was applied to CAP-D and UC, and both reports say that approval was conditional on specific approaches to IT procurement and software development being adopted (both derived from web development practice but arguably unproven for the specific contexts of these two projects).
"This separate but powerful influence was indicated by the reports to be the root cause of many problems in the projects."
It added: "Both project reports state or imply that there was a focus in the projects on the technical component, particularly the web front-end (a result of the ambition of 'digital transformation')."
Debates about IT procurement and development methods, such as waterfall vs agile, "appear to have been intense and contentious, and bled over into the dynamics of the projects as a whole," it said.
Repeating what the Public Accounts Committee has said on numerous IT projects, a failure of project ownership via a senior responsible owner caused "messed-up project governance due to misunderstanding and/or misallocation of key roles." ®