Why are big government IT projects still going wrong?

From digital exemplar to Whitehall fiasco

Then, in December, an NAO inquiry into the project revealed that the cost of the programme, initially slated at £154m, had spiralled 40% above the original budget to £215m of taxpayers’ money.

But what caused the programme, originally hailed as a “digital exemplar”, to turn into what PAC chair Meg Hillier labelled an “appalling Whitehall fiasco”?  

The PAC report put a lot of the blame on the GDS, saying its insistence on making the service a digital exemplar came at the expense of delivering something farmers actually could, and wanted to, use.

“The GDS introduced a level of innovation and risk to the programme, without assessing whether the department was capable of managing the changes, and did not provide sufficient support during implementation,” the report said, adding that the GDS had lost sight of the desired outcome, which was to pay farmers as quickly as possible.

The PAC report even suggested that the GDS had hindered the successful delivery of the programme.

But should the blame be put on the GDS? Although it is clear that the service’s focus on developing a digital front end did result in poor management of the programme in general, a lack of collaboration between the RPA, Defra and the GDS also contributed.

GDS chief technology officer Liam Maxwell’s comment to the PAC last December that rifts were caused by a culture clash and differences between the departments, which led to MP Richard Bacon calling him “Mr Fancypants”, didn’t do him any favours.

Neither did RPA chief Grimshaw’s comment that the emphasis on digital by default could have been a distraction from the RPA’s focus on delivering payments on time.

Different projects, same story

The RPA itself has a chequered history of delivering IT systems. In 2005, it began the project to introduce its £350m single payment scheme, which ended up costing more than four times the original estimate of £75m, with extra costs of £680m to administer the system.

A lack of clear priorities, changing leadership and the inability to publicly admit that a project is going wrong could be seen as a classic Whitehall combination, and the rural payments programme is by no means the only example.

In 2015, the NAO criticised the government’s e-Borders programme for its high level of turnover. The programme had a succession of five directors, including three interims, and between 2010 and 2014 there were two further programme directors.

The aim of the e-Borders programme was to improve the use of information to track people moving across UK borders by conducting checks on travellers at the start of their journey to the UK, rather than on arrival. It has cost £830m so far, with a further £275m in projected costs before the Home Office produces new systems in-house, which are due to go live in 2019.

Another GDS digital exemplar, the Department for Work and Pensions’ (DWP) universal credit programme, follows a similar pattern. In June last year, Computer Weekly reported that the total lifetime cost of the programme had increased by nearly £3bn in the previous three years to £15.8bn.

Why? According to an NAO report, it was due to a catalogue of failures in leadership and project management, and a culture of secrecy and ignoring warnings within the DWP.

And in the NHS, a report by the NAO revealed that the cost of the GP Extraction Service (GPES) had increased from £14m to £40m, with £5.5m of work being written off. 

GPES, which aims to collect data from the clinical systems of 8,000 GP practices in England, was originally due to start in 2010. The project has been heavily criticised by the PAC for its failure to learn from past mistakes.

“We keep calling for lessons to be learned and keep receiving reassurances from senior accounting officers that they are,” PAC chair Meg Hillier said at the time. “Yet the same issues occur time after time. It’s simply not good enough.”

Lack of skills

It is often said that the biggest challenge of digital transformation is culture change. But whose culture is it that needs to change? The government often emphasises transparency and its efforts to become one of the most transparent governments in the world. But when it comes to failure, it seems rather less willing to speak up.

Would the problems of major projects be solved more quickly, or even avoided altogether, if someone had the guts to put their hand up and say “this isn’t working” sooner?

An NAO report, published earlier this year, found that out of 149 major government projects with total lifetime costs of £511bn, including 40 ICT projects, one-third were rated red or amber-red, which means the successful delivery of the projects was in doubt or unachievable unless action was taken.

The report strongly criticised the government for its poor track record in delivering major projects, saying that despite the government taking steps to improve project management and capability, it was still difficult to tell “whether performance is improving” because there was no reliable and consistent measure of a project’s success.

Another NAO report revealed a serious skills gap that was affecting the government’s delivery of digital transformation programmes.

It said there were “not many digital and technology leaders in place”, and most of those had not been in their posts for long, with 73% of the digital and technology leaders surveyed having been in their job for less than two years.

Systemic failure

So, when it comes to delivering major projects, is the bar set too high at the start, failing to take the skills gap into account? It is a hard task to pinpoint who is to blame for what seems to be a systemic failure.

In the case of the rural payments service, the PAC recommends that Defra reviews its approach “to tackling serious failures of management and put in place measures to stop this ever happening again”.

For GDS, the PAC offers this advice: “The Cabinet Office, through its GDS, should comprehensively assess departments’ capabilities to deliver any changes it imposes and ensure that it provides an appropriate level of support for those changes.”

Is the government listening? The common expectation among the public now seems to be that any large government IT project will inevitably suffer delays and cost overruns, and start to resemble a shambles. It’s up to Whitehall to turn around that public perception.